Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Exploring urban change using historical maps: the industrialization of Long Island City (LIC), New York
(USC Thesis Other)
Exploring urban change using historical maps: the industrialization of Long Island City (LIC), New York
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
EXPLORING URBAN CHANGE USING HISTORICAL MAPS:
THE INDUSTRIALIZATION OF LONG ISLAND CITY (LIC), NEW YORK
by
Elizabeth J. Mamer
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) August 2015
2015 Elizabeth J Mamer
ii
DEDICATION
In memory of my father, who loved and collected old maps.
iii
ACKNOWLEDGMENTS
I would like to thank Karen Kemp for her continued guidance as I worked my way through this
research. Her patience and instruction were invaluable.
Thank you as well to my mom and sister for their constant support throughout this
process. A great thanks to Florence, whose boundless and persistent spirit motivated me. And
finally, to Spencer, for both encouraging me and putting up with me.
i
TABLE OF CONTENTS
DEDICATION ii
ACKNOWLEDGMENTS iii
LIST OF TABLES iii
LIST OF FIGURES iv
LIST OF ABBREVIATIONS vi
ABSTRACT vii
CHAPTER 1: INTRODUCTION 1
1.1 Motivation 2
1.2 Study Area 3
1.3 Research Goals 5
CHAPTER 2: BACKGROUND AND LITERATURE REVIEW 7
2.1 Historical GIS 7
2.2 Historical Narrative: The Urban Development of LIC 8
2.2.1 Pre-industrialization 9
2.2.2 Industrialization 10
2.2.3 Transportation Expansion 12
2.2.4 Residential Changes 14
2.3 Trends in Industrial Societies 15
CHAPTER 3: DEVELOPING THE DATASET 17
3.1 Historical Maps as Data Sources 17
3.2 Georeferencing 19
3.3 Digitization 22
3.4 Data Organization 26
3.5 Assign Shifts 30
CHAPTER 4: EXPLORING THE STORIES 33
4.1 Enumerating the Shifts 33
4.2 Visualizing the shifts spatially 34
4.3 Drawbacks of the Widespread Culturally-Focused Method 36
4.4 Diving Deeper 37
4.5 The Story at the Local Scale 40
4.6 Looking Backwards 45
ii
4.7 Identifying Other Types of Change 47
CHAPTER 5: DISCUSSION AND CONCLUSIONS 50
5.1 Implications and Limitations 50
5.2 Further Historical Conclusions 52
5.3 Future Work 53
5.3.1 Applying Methods Elsewhere 54
5.3.2 Expanding the Research 55
5.4 Summary 56
REFERENCES 57
APPENDIX A: Georeferenced Maps 60
APPENDIX B: ModelBuilder Model 65
iii
LIST OF TABLES
Table 1 Turnstile percentage change at subway stops in LIC and Astoria between 1920 and
1927................................................................................................................................... 15
Table 2 Map years and sources..................................................................................................... 19
Table 3. Building use type and subtype ........................................................................................ 24
Table 4 The count of each building use type for each year. ......................................................... 28
Table 5 Classification of shift type from originating year to the consequent year ....................... 30
Table 6 Shift ratings for each point type. 31
Table 7 Sample of the resulting dataset, documenting Types, Subtypes, Ratings and Shifts
through time periods ......................................................................................................... 32
Table 8 Each type of shift for each period.................................................................................... 33
Table 9 Subtypes included under the Other type in the new dataset ............................................ 40
Table 10 Number of each type of shift in Hunter’s Point............................................................. 40
Table 11 Percentage change in point types between each time period......................................... 41
Table 12 Shift rate in shifts per year............................................................................................. 42
Table 13 Initial building type and consequent building type through three time periods ............ 46
Table 14 The subtypes “Other” type and its shifts to industrial locations.................................... 47
iv
LIST OF FIGURES
Figure 1 Long Island City in Queens County, New York City. ..................................................... 1
Figure 2 Long Island City and Long Island (2015). ....................................................................... 4
Figure 3 1896 Long Island City by J.S. Kelsey .............................................................................. 5
Figure 4 Astoria and historically significant LIC sub-neighborhoods............................................ 9
Figure 5 Hunter’s Point, LIC. (left) 1849 (right) 1858 after the land had been leveled to
install streets...................................................................................................................... 10
Figure 6 Hunter's Point on a Sunday in 1898. .............................................................................. 11
Figure 7 Transportation methods and years in LIC ...................................................................... 13
Figure 8 Sanborn map from 1915, Sheet 16 ................................................................................. 18
Figure 9 1891 New York Atlas Map of LIC................................................................................. 19
Figure 10 The black and white key, showing all forty-seven Sanborn maps covering LIC......... 21
Figure 11 Mosaic of 1936 Sanborn maps georeferenced to modern-day LIC.............................. 22
Figure 12 Ravenswood Park lot changes from 1891 to 1950 ....................................................... 25
Figure 13 Type and subtype of cultural institutions for all five years. ......................................... 27
Figure 14 Cultural institution locations over all five years........................................................... 29
Figure 15 Shift types for each time period.................................................................................... 35
Figure 16 Diagram demonstrating how shifts were determined based on the footprint of
the latter year’s categorization.. ........................................................................................ 39
Figure 17 Building use shifts in Hunter’s Point from 1898 to 1950............................................. 43
Figure 18 Industrial Shifts and Residential Disappearances and their distributions over
three time periods.............................................................................................................. 48
Figure 19 The Atlas of New York 1891 Long Island City map georeferenced to modern LIC... 60
Figure 20 Sanborn 1898 Long Island City map georeferenced to modern-day LIC .................... 61
Figure 21 Sanborn 1915 Long Island City map georeferenced to modern-day LIC .................... 62
v
Figure 22 Sanborn 1936 Long Island City map georeferenced to modern-day LIC .................... 63
Figure 23 Sanborn 1950 Long Island City map georeferenced to modern-day LIC .................... 64
Figure 24 ModelBuilder model that organizes cultural institution point data. ............................. 65
Figure 25 ModelBuilder model to determine type of shifts in between time periods .................. 66
Figure 26 Python script assigning shift types based on previous year's category ........................ 67
vi
LIST OF ABBREVIATIONS
AAG Association of American Geographers
GIST Geographic Information Science and Technology
LIC Long Island City
LIRR Long Island Railroad
NYC New York City
RMS Root Mean Square
SSI Spatial Sciences Institute
USC University of Southern California
vii
ABSTRACT
The goal of this thesis was to develop a process in which historical land use can be tracked in
order to gain a better understanding of an area’s history. The study area, Long Island City (LIC) is historically an industrial neighborhood within Queens County of New York City. By
documenting its land use shifts from 1891 to 1950, it is possible to visualize and analyze the
changes that occurred as industrialization took place.
This study compiles a digital historical narrative to provide a foundation for
understanding the data, as well as a reference for making new conclusions from the results of the
analysis. Old fire insurance maps provide building footprints categorized by use. These were
used to digitize locations of interest as points that were catalogued under five different
categories: Cultural, Industrial, Residential, Shop, and Vacant at each of five time periods. The
resulting spatiotemporal database makes it possible to track a single building and its use through
a period of 59 years.
The methodology developed for this thesis collects and classifies building use as points
so as to develop efficiently and quickly an accurate historical dataset. In doing so, the project
tracked the cultural development of LIC through an examination of a set of key buildings, as
well as the overall land use change of a sub-neighborhood, Hunter’s Point. It determined that by
tracking the use of every building through every map year, one gets a better historical analysis.
Such methods can be used not only to help support previously known historical narratives, but
also to allow for new conclusions to be drawn.
1
CHAPTER 1: INTRODUCTION
Long Island City (LIC) is a neighborhood in southwestern Queens County, New York. It is the
closest neighborhood to Manhattan in Queens, making it particularly accessible by both public
transportation and car (Figure 1). However, unlike many outer-borough neighborhoods that are
close to Manhattan, LIC has not experienced the same public popularity. Instead, throughout
much of its history, it has undergone rapid and constant industrial business. This thesis enriches
the narrative of its industrialization through digital historical analysis.
Figure 1 Long Island City in Queens County, New York City.
2
LIC’s urban history began in the late 1850’s when farming activity was wiped out to
make room for industrial endeavors. By 1912, LIC had the largest number of factories and
factory employees in all of Queens (Seyfried 1984). When industry peaked in 1950, much of LIC
was covered in warehouses, factories, and refineries.
This study attempts to track the historical development of LIC using a method of
collecting, categorizing and organizing point locations of buildings through historical maps. By
tracking shifts in building use, the research creates a method for using historical maps to extract
data that can be investigated within a single, spatiotemporal database. The results of this process
are used to demonstrate LIC’s industrialization as it relates to its historical chronicle.
1.1 Motivation
In 1961, when manufacturing was beginning to decline, New York City rezoned Hunter’s
Point(a sub-neighborhood in the southwest portion of LIC) to encourage further industrialization
(Vitullo-Martin 2004). As manufacturing continued to wane in the 1970’s, the area maintained
the same zoning laws and was therefore, unable to adapt to the changing economy. Freight
services, factories and warehouses were abandoned, ultimately being subjected to graffiti and
vandalism.
Only within the last few decades has LIC started to move away from its industrial roots.
In 1993, New York City developed a new vision for LIC, entitled “Plan for Long Island City: A
Framework for Development.” This project established a list of goals for the decaying
neighborhood, including new parks and a new zoning strategy(NYC Planning 1993). Today,
LIC has developed a fresh, unique identity, combining a high-end look with its old industrial
flavor. Nevertheless, abandoned warehouses and factories still exist, demonstrating the deep
impact industry had on the neighborhood.
3
This study began with a curiosity about LIC’s current conundrum. Although it is
successfully moving away from its industrial past, it is apparent that industry had a lasting effect
on LIC, leading to the question, where and when did this deep impact originate? This study
tracks the process of industrialization in LIC by using fire insurance maps, which provide rich,
detailed evidence about land use and its changes through the years. It traces from the very
beginning of industrial endeavors in 1891, to the peak of industrialization, in 1950.
1.2 Study Area
This document refers to the study area as LIC, using the neighborhood’s current name
preference. On February 17, 2013, the New York Post published an article entitled, “‘Island’
nabe [sic]: Call us LIC!”. In this article, Gary Buiso, explains that Long Island City officials and
leaders want to change the neighborhood’s name to simply LIC. Buiso quotes the head of the
Queens Local Development Corp. Tourism Council, Rob MacKay, as saying, “It puts us out on
Long Island, and that’s inaccurate—we are urban and hip”(Buiso 2013, para. 4). While Long
Island City is the closest neighborhood to Manhattan in Queens, its name implies that is a part of
Long Island, an area that is not only connected to the opposite side of Queens, but that is also not
a part of New York City (Figure 2). According to these community members, such confusion
poorly affects LIC’s economy and growth.
4
Figure 2 Long Island City and Long Island (2015) Moreover, this study uses the name LIC so as to differentiate between the historical
boundaries of Long Island City and the modern-day boundaries; the latter being the extent of this
study. Despite its name, modern day LIC (Long Island City) is not a city. Before Queens was
incorporated under New York City in 1898, Long Island City was, in fact, a city unto its own and
it covered the whole western shoreline of Queens(Figure 3). The northern part of historical Long
Island City is now a separate neighborhood known as Astoria, but in the late 19
th
century,
Astoria was simply known as the residential portion of greater Long Island City (Seyfried 1984).
Consequently, the name LIC not only reflects the recent initiative to improve the neighborhood,
but also the modern-day boundaries that distinguish it as a neighborhood as opposed to the city it
once was. This study will refer to the neighborhood as LIC so as to identify its modern
boundaries, as seen in Figure 1.
5
Figure 3 1896 Long Island City. Retrieved from “History of Long Island City, New York”
by J.S. Kelsey
1.3 Research Goals
The purpose of this study was to create a process for documenting urban change using
historical maps in order to track LIC’s industrialization and its effect on the neighborhood. As a
means of selectively mining the data within a large study area, the study began by focusing on
the decline of cultural and community institutions as an indicator of industrial growth. The initial
hypothesis was that cultural institutions were replaced in the landscape with industrial factories.
This project looked to show that LIC became an industrial neighborhood, moving social and
cultural conveniences, such as schools and parks, to other, nearby neighborhoods.
6
The project began by documenting the evolution of LIC cultural institutions through five
different years of historical maps. Since cultural institutions are reflective of a community’s
social capital (that which promotes civic engagement), they are quickly affected when industry
permeates a neighborhood (Putnam 1993; Mohan & Mohan 2002; Oulton 2012). Consequently,
by tracking changes in these cultural locations it should be possible to identify neighborhood
changes. The results of the application of this method to LIC were analyzed to assess its overall
efficacy in documenting general historical change.
The initial assessment revealed that further exploration was needed. LIC’s
industrialization was not explicitly apparent within this method, as many cultural institutions
persisted. To better understand the neighborhood’s development, the same method of tracking
building use was applied to every building—regardless of type—at a smaller, local level. The
result provides a more generalized insight into both the industrialization process, as well as the
overall urban history of LIC.
This thesis begins by establishing a historical narrative as a foundation for the data. It
then delves into the methodology for locating historical maps, georeferencing them, extracting
the data, and finally, organizing those data. Following this, it reports on the analysis and
manipulation of the data so as to put them in the context of the historical narrative. It compares
the results with the recorded events throughout the 59 years to both prove and expand upon the
neighborhood’s story. As mentioned, while the first method proves insufficient, the second
method shows the full process of industrialization and its effects on other building uses
throughout time. This thesis ultimately corrects, falsifies and enhances the historical narrative of
LIC.
7
CHAPTER 2: BACKGROUND AND LITERATURE REVIEW
Using GIS to understand industrial development in history brings a scientific process into more
conceptual disciplines (Bodenhamer, 2010). To gain a deeper understanding of theoretical
concepts, GIS establishes a spatial setting from which one can determine temporal patterns
(Kemp, 2009). It gives context to complex phenomena that are often hidden within abstract texts
or even empirical data. This chapter delves into both the texts that have documented LIC’s
history, and the texts that study spatial industrialization so as to provide a background for a GIS
analysis.
2.1 Historical GIS
Ian Gregory asserts that there are three concepts that make up GIS: attribute, space and
time (2010). Although many GIS studies are temporally stationary, the attributes and space are a
reflection of a moment in time. Historical GIS manifests the dimension of time, showing both the
cause and effect of time’s events on attributes and space.
Combining history and GIS, however, can be a tricky process due to the divergence of
the two disciplines. Historians are hesitant to use quantitative approaches and cartographers
struggle with the representation of uncertain and ambiguous data in a thorough manner (Knowles
2008). While historians typically study the conceptual factors of time, GIS is often used to
analyze data at either stagnant or timed moments. When trying to merge these two areas of study,
it can be difficult to maintain the accuracy of normal data manipulation within the theoretical
field that is history (Gregory 2010).
Another obstacle of Historical GIS is the myriad of “stories” that arise when combining
both space and time (Massey 2013). Neither space, nor time, are motionless and to study both
8
together brings a, “dimension of multiplicity” in which stories can be never-ending(Massey
2013, par. 5). Add Gregory’s third Historical GIS concept, attribute, and the stories grow further.
Although the breadth of these stories can be overwhelming and difficult to organize, they
are also what make Historical GIS so useful. Many historians study history using their own
“mental map” of the area, documenting and analyzing events within their own spatial
interpretations(Lynch 1960; Pascoe 2010). Historical GIS explores these interpretations by
actualizing the mental maps and delving into their myriad of stories. The resulting data can speak
to these stories, ultimately substantiating or disproving them.
In order to tackle the effects of such complexity, this study begins by referencing several
sources to establish a particular mental map of the neighborhood. It creates a singular and linear
historical narrative of LIC, focusing on urban development. Forming a chronicle provides
context and is an essential back-drop for representing the data(Raymond 2011). In this way, it is
possible to elucidate the mental map to determine its accuracy. Since this study looks to record
the development of an industrializing city, the historical narrative documents this aspect of LIC’s
history.
2.2 Historical Narrative: The Urban Development of LIC
A historical narrative of LIC provides a framework for this study, which helps to identify
patterns in building use throughout time. Just as the study is focusing on LIC’s urban
development, so does the historical narrative. This section chronicles political, social, and
planning changes that occurred throughout the 59 year period during which the study takes place.
This research helps to examine whether the events of the narrative had an effect on the
neighborhood by providing context for the results.
9
Figure 4 indicates the various sub-neighborhoods of LIC, as well as the neighborhood to
the north, Astoria. While these boundaries are definitive, the sub-neighborhoods are important to
the history of LIC and are frequently referenced within the historical resources.
Figure 4 Astoria and historically significant LIC sub-neighborhoods
2.2.1 Pre-industrialization
The greater Long Island City—that now consists of LIC and Astoria as seen in Figure 3—
emerged in the 1630’s, when the Dutch at New Amsterdam awarded a 160 acre land grant in the
northwest peninsula of Astoria(Queens West Villager 2011). Throughout the following decades,
this area, known as Jarck’s Farm, switched ownership back and forth between the Dutch and the
Native Americans. Nevertheless, this settlement marked the first Long Island City community.
10
LIC, as it is known today and pictured in Figure 1, was established by George Hunter in
the early 1800’s, supplying the southwest neighborhood with its current name, Hunter’s Point.
When he and his wife died, their sons sold the 210 acre estate to Reverend Eliphelet Nott for
$200,000 (Seyfried 1984). Dr. Nott and his friend, Neziah Bliss, an investor in steamboats and
mills, built up the land together. In 1852, they leveled the estate and the underlying hills and
installed building lots for homes and factories(Figure 5).
Figure 5 Hunter’s Point, LIC. (left) 1849 (right) 1858 after the land had been leveled to
install streets. Retrieved from “300 Years of Long Island City” by Vincent Seyfried (1984) 2.2.2 Industrialization
Following the landscape demolition, a ferry service opened, connecting Hunter’s Point to
east 34
th
Street in Manhattan. Nott and Bliss also started working with the Long Island Railroad
(LIRR) to create a train station that would service more easterly villages in Queens. They won
the bid and a terminus station opened up at 54
th
Avenue and 5
th
Street of Hunter’s Point in 1854
11
(Kelsey 1896). These improved transportation methods signified the start of LIC’s urban
development. Seyfried explained, “industry now became possible for the first time with the
railroad facilities of an island-wide railroad with its own boats to bring in raw materials and to
take out finished products” (Seyfried 1984, p. 87).
At this point, farming had disappeared from LIC, leaving plenty of open space for
developers. Furthermore, as transportation methods continued to progress, more companies
began bringing their business to LIC. Theodore and William Steinway were two influential
brothers, who opened up a piano factory in LIC in 1869. Over time they profited so much from
their piano manufacturing endeavor that in 1883, they opened up their own railroad company,
Hunter’s Point Railroad Company, allowing better access to and from LIC for their workers.
(Queens West Villager 2011). By the late 19
th
century, ferries and railroads all funneled
passengers directly to Hunter’s Point, creating a commercial, industrial and residential hub
(Figure 6).
Figure 6 Hunter's Point on a Sunday in 1898. Retrieved from “300 Years of Long Island
City” by Vincent Seyfried (1984)
12
In other parts of LIC, industrial endeavors expanded. Newtown Creek became a busy
waterway as many bridges were torn down to install drawbridges that could allow large boats to
pass through. In Ravenswood, many of the old mansions and parks that lined the shoreline were
abandoned and demolished. Of this degradation, an 1894 New York Times article entitled “Old
Long Island Mansions” wrote:
Old mansions, with tales to tell, line the East River front in Ravenswood, Long
Island and City. Separated by extensive grounds, which were once well-kept
parks, these relics of past grandeur stretch along Vernon Avenue, from sooty
Hunter’s Point quite into Astoria. Not more than twenty years ago famous
families of that period filled these great houses with life and fashion. Black clouds
of smoke now hang over these once beautiful homes, which are streaked and
seamed. The carved stone and iron fences have been demolished, the grounds laid
waste… Manufactories and other industries gradually drove nearly all of the old-
time residents out of their great houses.(New York Times(unauthored 1894,
para. 1) Steadily LIC became a largely industrial neighborhood; factories and their smoke filled the
skyline.
The next big turning point for LIC occurred at the end of the 19
th
century, when Queens
became administratively apart of New York City. On May 11, 1896, Governor Morton signed
the bill, leading to the end of the local government on December 31, 1987. Once the local
politicians learned that they would have no accountability, they started granting franchise and
construction contracts to friends. Consequently, when Long Island City was officially
consolidated into New York City in 1898, it had a great amount of construction, but had acquired
massive amounts of debt(Seyfried 1984).
2.2.3 Transportation Expansion
After LIC was consolidated, various transportation projects were put into motion in order
to connect Queens to Manhattan (Figure 7). In March 1909, Queensborough Bridge opened,
connecting 59
th
Street in Manhattan to LIC. To the northwest of Hunter’s Point, Sunswick
13
Meadow was filled in to make room for the end of the bridge. This area, now known as Queens
Plaza, quickly became a transit hub. When World War I began in 1914, even more factories and
industrial plants opened in LIC. Newtown Creek, the southern portion in Figure 4, experienced a
sharp industrial rise as factories and refineries quickly settled along the shore. Such growth
meant a need for transportation for commuters to get to and from work. In 1914, the
Pennsylvania tunnels opened at the Hunterspoint Avenue Station, bringing passengers from Penn
Station to LIC. Shortly thereafter, in 1915, the Steinway tunnels opened, transporting passengers
from Manhattan to the Jackson-Vernon Station in Hunter’s Point. Over the next few years, the
train continued north, ultimately connecting with the bridge and Queens Plaza.
Figure 7 Transportation methods and years in LIC
14
The Queensborough Bridge, Pennsylvania tunnels, and Steinway tunnels all greatly
affected Hunter’s Point. From 1909 to 1915, the bridge traffic to and from Queens Plaza grew
more than 325%. Conversely, traffic to the Hunter’s Point LIRR station decreased by 77%
(Seyfried 1984). Many of the commuters who had once used the ferries and LIRR to get to LIC,
now began using these other methods, both of which were not located in Hunter’s Point.
Consequently, the ferries closed in 1925 and downtown Hunter’s Point experienced a gradual
decline.
2.2.4 Residential Changes
After World War I, Astoria experienced a great residential boom. Vincent Seyfried
attributed this to three main causes: young families formed from the war, searching for
inexpensive housing, new investors who had profited from the war looking for real estate, and
finally, LIC workers, who no longer wanted to commute all the way from Manhattan or
Brooklyn, searching for housing closer to their jobs(Seyfried 1984). LIC was already full with
industrial and transportation construction, but Astoria still had many open areas for new
construction. Consequently, not only did many Brooklyn and Manhattan residents move to
Astoria, but so did LIC residents, looking to trade LIC’s industrial character, for Astoria’s
residential one.
Astoria quickly became a busy residential neighborhood whose inhabitants commuted to
many other parts of the city. Table 1 describes the number of people commuting through the
turnstiles from the first subway stop in LIC (39
th
Street), up into the subway stops in Astoria
(36
th
, Broadway and 30
th
Avenues) from 1920 to 1929. The statistics show a drastic increase in
commuters at the three stops within Astoria, while the first and only stop in LIC increases
steadily, but not with the same extreme increments.
15
Table 1 Turnstile percentage change at subway stops in LIC and Astoria between
1920 and 1927
Stop Neighborhood 1920 1920 to 1927 1927 to 1928 1928 to 1929
39
th
Avenue LIC 790,360 56.01% 2.72% 8.31%
36
th
Avenue Astoria 1,026,600 91.75% 9.48% 20.12%
Broadway Astoria 2,063,800 101.05% 12.33% 12.28%
30
th
Avenue Astoria 2,044,800 106.85% 11.11% 8.78%
Source: Data adapted from Seyfried’s “300 Years of Long Island City” (1984 p. 155).
The stock market crash in 1929 ended the housing boom of the 1920’s. Queens, as a
whole, felt the effects of the Great Depression, however LIC’s already industrial character
endured. Both the 1939 World’s Fair in Flushing Meadows, Queens and the build up to the
World War II, kept LIC’s manufacturing business afloat. These industrial practices remained
consistent throughout the war(Asadorian and Seyfried 1991) Manufacturing in the United States remained steady throughout the first half of the 20
th
century and peaked in 1950. However, between 1973 and 1975, production decreased by almost
12% (Federal Reserve Bank of New York Annual Report 1976). LIC experienced much of this
deterioration as it had long relied on its industrial character. As factories were abandoned and
unemployment soared, LIC experienced a sharp economic decline in the second half of the 20
th
century.
2.3 Trends in Industrial Societies
Industrialization has a dramatic effect on culture in a city. As production increases, so
does the population’s income. Such economic growth allows for investment in social capital
within the community (Mahdavi and Azizmohammadlou 2013). Consequently, industrialization
can help to develop cultural institutions, such as schools and churches, which help benefit the
sense of community. For much of the 19
th
century, LIC experienced these effects. The ferries
16
brought people from Manhattan to Hunter’s Point where saloons, shops and homes were
abundant.
However, when Long Island City was consolidated under New York City in 1898, it was
no longer a city unto itself. It swiftly became a neighborhood within a much larger city. Despite
nurturing economic and cultural development, industrialization can also trigger separation of
land uses as dense cultural areas become separated from the industrial neighborhoods where
people commute to and from (Gilliland and Olson 2013; Mahdavi and Azizmohammadlou 2013;
Pratt 1911). This was especially noticeable for LIC as a new neighborhood in the early 1900’s.
Rather than live in LIC, factory workers commuted from other parts of New York City in order
to work there and consequently, the neighborhood no longer needed to provide the same amount
of cultural and social services that a whole city would need (Pratt 1911).
By tracking the late 19
th
and early 20
th
century development of LIC, one would expect to
see the effects of such segregation on cultural institutions. Knowing both these industrial trends,
as well as the events established in the historical narrative, provides context for this study’s data.
Ultimately, this thesis creates a method for collecting building points and their uses from
historical maps so as to identify and analyze both industrial trends and historical events. The
resulting data is compared with this foundational research in order to support it, as well as
identify new patterns.
17
CHAPTER 3: DEVELOPING THE DATASET
Having established the historical narrative of the land and time, it is possible to turn to historical
maps to provide context. Prior to GIS technology, illustrated maps were a great way of
documenting and organizing information about the land. However, such maps were more
qualitative than quantitative, since they did not have a proper way of discerning the intricate
pieces of data. Lack of quantitative data in these maps, makes it difficult to follow changes in the
land from one year to the next. This chapter discusses methods that can be used to find, collect
and organize data in historical maps as a basis for quantitative analysis. It specifically examines
historical LIC maps to compare their illustration of the neighborhoods to the previously
examined historical narrative. In order to use GIS to document change in these maps, it was
necessary to first gather and classify data from them so that the relevant information was
available and highlighted. This process required five main steps: locating relevant historical
maps, georeferencing maps, digitizing features, organizing data, and assigning shift types. The
culmination of these steps ultimately helped to confirm the historical narrative, and subsequently,
expand upon it.
3.1 Historical Maps as Data Sources
Sanborn maps are fire insurance maps that document buildings in cities all over the
United States. Produced from 1867 to present day, these maps provide snapshots in time that
illustrate existing building footprints, construction details and building uses, as well as street
names and water lines(Figure 8). Sanborn maps were this study’s primary resource for
documenting historical LIC, as they provide a comprehensive record of the location and
functions of buildings throughout the period of interest. Such detail made it possible to track the
building uses across several years in order to detect change in the neighborhood’s history.
18
Figure 8 Sanborn map from 1915, Sheet 16
The Atlas of New York also produced maps similar to Sanborn maps. They too
documented land use by parcel, however, their maps covered a larger extent of land. Whereas
each Sanborn map only covers four to eight blocks, The Atlas of New York Maps cover over a
hundred. Furthermore, digitally acquired Atlas of New York Maps tend to appear in color
(Figure 9). This is due to the fact that they were directly scanned, whereas the Sanborn maps
were scanned from microfilm. Since the earliest Sanborn map of LIC is 1898, this study also
used an Atlas of New York map from 1891. This allowed for the study period to reach further
back in time.
19
Figure 9 1891 New York Atlas Map of LIC
Ultimately, this study examined a total of five time periods using The Atlas of New York
and Sanborn maps(Table 2). The maps spanned from 1891, when LIC had begun to move away
from its farming past, to 1950, when LIC was at its peak of industrial production (Greenberg
2008). This time span allowed for a comprehensive evaluation of an evolving industrial area and
its effects on the cultural aspects of the neighborhood.
Table 2 Map years and sources
Year Company
1891 The Atlas of New York
1898 Sanborn Maps
1915 Sanborn Maps
1936 Sanborn Maps
1950 Sanborn Maps
3.2 Georeferencing
To begin the process of documenting the neighborhood’s evolution, it was necessary to
first georeference the raster maps. Consequently, these maps were aligned to modern-day LIC
20
under coordinate system, NAD 1983 State Plane New York Long Island. Intersections between
streets provided the control points for all the maps. Although the latest maps (1950, 1936, and
1915) included modern-day street names, the earlier maps (1891 and 1898) did not. The 1915
Sanborn series proved to be a useful reference for these older maps, as it recorded the modern-
day streets, but also included their former names in parentheses. This information helped to
ensure that all control points were linking the correct historical intersections to modern-day
streets.
While the 1891 Atlas of New York covered the entire study area of LIC, it took forty-
seven Sanborn maps per year to cover the same extent (Figure 10). The Atlas of New York map
required a total of fifteen control points, for a root mean square(RMS) error of 5.17 under the
third polynomial. This indicates that there was approximately a five meter residual between the
control points.
21
Figure 10 The black and white key, showing all forty-seven Sanborn maps covering LIC
Each of the 47 Sanborn maps required five control points, but the shoreline maps did not
always meet this requirement. Since shorelines can change drastically over time, it was
impossible to use any portion of it as a reference point(USGS 2011). Consequently, many of the
shoreline Sanborn maps only had three or four control points, making their RMS errors greater.
Ultimately, this was not a serious source of error. Since the maps were referenced and digitized
according to modern-day LIC as a baseline, it was more important that the maps lined up with
one another. Therefore, when the next process of feature digitization began, each feature simply
needed to align with the location of its historical counterparts.
In order for quick rendering, the forty-seven Sanborn maps from each year were placed
into mosaics, the results of which are provided in Appendix A. This ensured quick image
22
rendering, but it also highlighted the areas of each individual map that had overlapping blank
map margins. Since the images were not trimmed prior to georeferencing, the borders of each of
the 47 maps overlapped one another(Figure 11). This ultimately made it difficult to view certain
portions of the study area on the mosaic, particularly when trying to digitize attributes that
appeared in the white areas. Nevertheless, when white space occurred in an area that needed to
be digitized, the corresponding original map was inserted to overlap the white space.
Figure 11 Mosaic of 1936 Sanborn maps georeferenced to modern-day LIC. (left) Each
footprint highlighting all forty-seven maps. (right) A portion of the mosaic without
footprints showing overlapping white spaces and mapsheet edges.
3.3 Digitization
To document and measure culture in an industrial society, it was first necessary to define
it. The classification of cultural points incorporated both available historical spatial data, as well
as previous literature on cultural institutions. In his essay, “Managing the unmanageable: The
politics of cultural planning”, Cliver Gray argues that culture can be categorized under two
categories: material and valuative (Gray 2004). Material culture, according to Gray, is formed
from resources and activities, such as playgrounds and recreation, while valuative culture is
formed from a society’s makeup, such as the people’s ethnicity, religion, and background. In this
case, this project evaluated culture from a material point of view, as this is the data that was
23
present in the historical maps. By examining these maps, it was possible to locate and digitize
any illustrated institutions that constitute material culture as defined by Gray. Ultimately, each
cultural category was defined first by Gray’s definition, and secondarily according to data
available on the historical maps as building labels and names. Consequently, material culture fell
into one of five physical categories: Religious Center, Museum/Library, School, Social Service
(such as Police Stations and YMCA associations), or Park.
Fire insurance maps contain an enormous amount of information. Digitally capturing and
categorizing all of that information comprehensively would take a great deal of time. In an effort
to develop a methodology that would be both effective yet efficient, only the locations of
buildings housing four of these five cultural categories—excluding parks—were digitized as
points on each of the five maps. By documenting these civic resources(places of material
culture) as points in each map year, it was possible to track their presence—or lack thereof—
over time while not worrying about matching evolving building footprints. Limiting the
digitization effort to only points at cultural locations significantly reduced the magnitude of the
digitization task while still allowing the study question to be addressed. Related studies may
ultimately benefit from documenting every building within the study area as a point, and
analyzing its change throughout the years, but with such a large study area, the process would be
onerous. This approach, however, is time-effective and puts a greater emphasis on the cultural
points and their shifts, ultimately reducing—albeit limiting—the data.
To record these building use shifts, each digitized point that referenced a cultural
institution in at least one year, needed to be classified for all five years. For instance, a building
that was a school in 1915, was also classified by use for all other four years: 1891, 1898, 1936
and 1950. Sanborn maps meticulously categorize building use based on “size, shape, and
24
construction of dwellings, commercial buildings and factories,” as well as providing “widths and
names of streets, property boundaries, building use, and house and block numbers.” (Ristow
2014, para. 1) Such details were used to determine the use of a building as it evolved through
time. This study generalized the Sanborn building use types into five categories: Cultural, Shop,
Residential, Vacant, and Industrial. The Cultural type was then given the previously stated
cultural institution subtypes (Table 3).
Table 3. Building use type and subtype
Type Subtype
Cultural
School
Religious Center
Social Service
Museum/Library
Park
Shop Shop
Residential Residential
Vacant Vacant
Industrial Industrial
While points represented religious centers, museums/libraries, schools and social
services, polygons represented parks. This is due to the fact that the study is not about building
use footprint, but the building use presence. Consequently, the first four categories were
digitized as points to emphasize quantity and magnitude of building use shift over time. Parks are
the exception to this rule, as they hold a bigger footprint and can be overtaken throughout time
by multiple buildings. This study focused on point data to understand cultural shifts, but also
digitized park polygons in order to have a complete visualization of the area’s cultural
development. For instance, Figure 12 shows a close up view of the changes that occurred over
time to Ravenswood Park, a large park that resided in the center of LIC in 1891. By simply
25
digitizing and classifying the Sanborn and Atlas of New York maps, it was possible to observe
the conversion from an initial cultural institution, to an area occupied by factories.
Figure 12 Ravenswood Park lot changes from 1891 to 1950
26
3.4 Data Organization
Once all the points were digitized, it was necessary to organize them. Since each point
was digitized according to individual historical maps, all the points were in separate feature
classes corresponding to their year. However, in order to grasp how these points had shifted from
one year to the next, it was necessary to combine them. In the essay, “Denny Regrade, 1893-
2008: A Case Study in Historical GIS”, Aaron Raymond explains that this approach to
organizing a historical dataset allows for a comprehensive analysis of both the feature’s presence
at a singular moment in time, as well as its existence from year to year (Raymond 2011).
To organize the point data once it was digitized, a ModelBuilder model (provided in
Appendix B) streamlined the process of merging, arranging, and assigning ID numbers to
digitized cultural institutions. It accomplished this by gathering all points that were once cultural
institutions into a single file and deleting any duplicates that indicated points that had remained
or reoccurred throughout the years. Then it assigned each feature an ID and copied them into five
additional feature classes, one for each year. It then assigned each class a year and combined the
features into one complete dataset by appending all years’ feature classes back into the first
year’s feature class. This model ensured that all cultural features identified on every map were
documented, had a point location ID, and included a year.
Such categorization readied the data to be organized by shift type, but also allowed for
preliminary analyses of building use types and placement. For instance, Figure 13 illustrates the
cultural institution subtypes for each year. While the data does not explicitly show the shifts in
points from year to year, it does give an introductory look into the types of cultural institutions. It
begins to give context to the historical narrative that was discussed earlier in Chapter 2 and it
starts to point out new stories. For instance, the Northwest corner slowly loses cultural
27
institutions, both religious centers and parks. This could possibly be attributed to the
development of factories along the shoreline, as described in Chapter 2.
Figure 13 Type and subtype of cultural institutions for all five years.
28
Organizing and categorizing the data in this way also allowed for a preliminary
examination of all building types for each point from year to year. Table 4 shows the number of
each building use type over all five years. Furthermore, Figure 14 illustrates the use type of
points that had once been a cultural institution. A visual analysis of this is used to understand
how they changed throughout the years. Table 4 shows that next to cultural institution, which
will predominate due to the nature of the data collection, vacant and industrial uses were most
prevalent at other times. Again, by considering the historical narrative, one could conclude that
the vacancies decreased as the area was built up, and industrial points increased as
manufacturing intensified. These types of conclusions, as well as the stories that can be
uncovered from the data, are discussed further in Chapter 4.
Table 4 The count of each building use type for each year.
Type 1891 1898 1915 1936 1950
Cultural 17 24 23 24 23
Industrial 2 0 3 8 14
Residential 5 6 3 1 0
Shop 1 3 7 4 2
Vacant 16 8 5 4 2
Although these data and maps are useful as a basis for understanding building use at
these individual points from year to year, it is difficult to visualize change. One can perhaps
review the data to get a static understanding of the points, but it is nearly impossible to
understand the changes temporally. Furthermore, the spatial pattern of change cannot be clearly
tracked. To assess these changes, this study took the point data and analyzed how they shifted
from one map to the next.
29
Figure 14 Cultural institution locations over all five years
30
3.5 Assign Shifts
To incorporate the aspect of time, these points were compared to one another in a forward
and linear context. Shifts between points recorded the type of change that occurred at that
location from one map year to the next. Consequently, there were four periods of shifts: 1891-
1898, 1898-1915, 1915-1936 and 1936-1950. The shift types were classified by comparing the
point of the originating year with the consequent year. If the latter year did not match the
originating year type, then the shift type reflected the building use type of the latter year.
Conversely, if it did match, then the shift type would indicate that there was no change in
building use type. This produced ten different shift types (Table 5).
Table 5 Classification of shift type from originating year to the consequent year
Originating
Year Type
Consequent
Year Type
Shift Type
Other
Cultural Cultural
Shop Shop
Residential Residential
Vacant Vacant
Industrial Industrial
Cultural Cultural Cultural No Change
Shop Shop Shop No Change
Residential Residential Residential No Change
Vacant Vacant Vacant No Change
Industrial Industrial Industrial No Change
A ModelBuilder model helped to simplify the process of establishing shift types. This
model is illustrated in Appendix B. It started out by assigning ratings for the type of point (Table
6). Then, using Calculate Field, a Python script added the previous point’s value to the current
feature’s value.
31
Table 6 Shift ratings for each point type.
Point Type Rating
Cultural 5
Residential 4
Shop 3
Vacant 2
Industrial 1
Another Python Script then used an If-Else statement to detect when there was no change
in building use type. It did this by identifying when the last feature’s rating was the same as the
current feature’s rating. For instance, if the “Shift Rating”(sum) was 6 and the current feature’s
Rating was 3, then that feature had remained a shop from one year to the next. All other Ratings
with a value of 3 indicated that the point had shifted to a shop, regardless of the previous year’s
category. Finally, every fifth entry was deleted, since this row incorrectly summed ratings for
different locations that were simply adjacent in the list.
These two ModelBuilder models produced a dataset containing four periods of shifts.
Each entry represents a point and documents how it has shifted from one year to another. Table 7
provides a sample of the dataset, showing the points’ ID, type, year, rating and sum of rating
from the previous year, and finally, type of shift that occurred.
Ultimately this data preparation provided the foundation for exploring the stories of
building uses in time and place. Aided by the historical narrative, the data gives context to the
development of the land, and even illustrates stories that may not otherwise be apparent. The
stories uncovered by the spatial and numerical analysis are explored in the following chapter.
32
Table 7 Sample of the resulting dataset, documenting Types, Subtypes, Ratings and Shifts
through time periods
OBJECTID Time ID Subtype Type Rating Shift
Rating
Shift Type
149 1891-
1898
39 Residential Residential 4 8 Residential
No Change
150 1898-
1915
39 Police
Station
Cultural 5 9 Cultural
151 1915-
1936
39 Police
Station
Cultural 5 10 Cultural
No Change
152 1936-
1950
39 Police
Station
Cultural 5 10 Cultural
No Change
153 1891-
1898
40 Library Cultural 5 7 Cultural
154 1898-
1915
40 Shop Shop 3 8 Shop
155 1915-
1936
40 Shop Shop 3 6 Shop No
Change
156 1936-
1950
40 Factory Industrial 1 4 Industrial
157 1891-
1898
41 YMCA Cultural 5 10 Cultural
No Change
158 1898-
1915
41 YMCA Cultural 5 10 Cultural
No Change
159 1915-
1936
41 YMCA Cultural 5 10 Cultural
No Change
160 1936-
1950
41 Factory Industrial 1 6 Industrial
161 1891-
1898
43 Residential Residential 4 9 Residential
No Change
162 1898-
1915
43 Residential Residential 4 8 Residential
No Change
163 1915-
1936
43 Industrial Industrial 1 5 Industrial
164 1936-
1950
43 Industrial Industrial 1 2 Same
Industrial
33
CHAPTER 4: EXPLORING THE STORIES
This method of data collection and organization produced a dataset of cultural points and their
transitions over time. The historical narrative developed in Chapter 2 provided background for
the data collection, laying a foundation through which to understand it. By using the extracted
data to further examine the historical narrative, it was possible to see if the data was in fact
telling the same story, or if additional analysis was needed. This chapter explores the dataset, as
well as its shortcomings in order to show its insufficiency. It then describes the use of the same
data collection process to develop a more robust dataset and explores the historical implications
as it relates to the narrative.
4.1 Enumerating the Shifts
Although some locations remain in the same category from year to year, many changed.
Table 8 lists the total number of each kind of shift in each year. There are naturally more cultural
shifts because cultural points were used as the basis for finding points, but there are other
conclusions that can be drawn from this data.
Table 8 Each type of shift for each period
Change to: 1891-1898 1898-1915 1915-1936 1936-1950
Cultural 8 8 7 2
Industrial 0 3 6 6
Residential 2 0 1 0
Shop 2 5 0 0
Vacant 1 2 2 0
One of the most noticeable changes in the data is the reduction of cultural shifts and
increase in industrial shifts as time moves forward. By 1950, industrial shifts surpassed the
cultural ones, meaning that more buildings had been industrialized than any other type of shift.
This is a contrast to the 1891 to 1898 time period, when there were eight cultural shifts and zero
34
industrial shifts. However, while these changes are noticeable, the numbers are not particularly
significant. Due to the nature of this data collection, cultural institutions will tend to have larger
numbers, making the comparison to industrial shifts negligible.
4.2 Visualizing the shifts spatially
More still can be uncovered by exploring the points’ shifts spatially. Such visualizations
not only back up the tabular results, but they also allow one to identify and track spatial trends
and relationships that may not have otherwise been noted (Knowles 2008). Figure 15 illuminates
these shifts, including both the points that have remained the same and the points that have
shifted to a different category. Similar to the tabular data, it is evident by viewing the maps that
there were no industrial shifts at these cultural points from 1891 to 1898, but industrial shifts
increased as time continued. By the 1915 to 1936 time period, there were many apparent
industrial shifts throughout LIC.
Throughout all four time periods, there were two areas that appeared to have the strongest
cluster of points; the southwest section, known as Hunter’s Point, and the northeast area, known
as Queens Plaza. These are the same areas that are historically significant for Long Island City
because they were both major centers of transit at different points in time. As discussed in
Chapter 2, in the late 19
th
and early 20
th
centuries, Hunter’s Point was considered the downtown
(Stadler 2014). However, when the Queensborough bridge opened in 1908, and the Long Island
Railroad (LIRR) extended into LIC in 1915, Queens Plaza became the new bustling center, as
Hunter’s Point lost many of its commuters. By July 1918, the number of commuters to Hunter’s
Point had fallen over 99% and in 1925, the ferries traveling from Manhattan to Borden Avenue
in LIC were shut down (Seyfried 1984).
35
Figure 15 Shift types for each time period
36
4.3 Drawbacks of the Widespread Culturally-Focused Method
Although industrial shifts appeared to increase in LIC over time, the hypothesis that loss
of cultural institutions would indicate industrialization did not hold true. In reviewing the data
and the map, many cultural institutions persisted through time. For instance, out of the twenty-
three cultural locations in 1950, only two had shifted from another type. The other twenty-one
had remained cultural institutions from the previous map year. Such results differ from the
hypothesis, as they show that many cultural institutions were not altered by the industrialization
of LIC, but were instead grounded in their existence.
Thus the chosen methodology not only appears to disprove the hypothesis, but it also did
not corroborate the complete historical narrative of LIC. The decision to focus only on collecting
cultural institution points from the historical maps was made deliberately to control the amount
of data that had to be extracted, making the effort more efficient and manageable, while still
acquiring the key data about the locations of cultural shifts. However, reducing the data also
meant limiting it, and in not documenting all the building uses through all of these time slices,
the story was incomplete.
In “Denny Regrade, 1893-2008: A Case Study in Historical GIS”, Aaron Raymond
asserts that there are three different data validation techniques for historical GIS: continuing data
validation during and after digitizing elements, verifying datasets against source material, and
testing datasets with sample selection queries (Raymond 2011). This study followed all three
techniques, however, upon executing the second, it was apparent that the data did not fully
support or enhance the historical narrative that had been created. There were certainly trends that
could be visualized from the current set of data, and it began to support the historical narrative,
but many of the more specific stories were not apparent. For instance, the historical narrative
suggests that as Queens Plaza was built up through several transportation initiatives, Hunter’s
37
Point began to decline. However, these maps do not indicate significant transformations in either
of these two sub-neighborhoods. Ultimately, this method helped to understand in what time
period cultural points shifted and what they changed into, but it was not useful in introducing a
truly detailed historical analysis of LIC.
Furthermore, as Ian Muehlenhaus explains, “It is particularly important to note that if a
sample is not randomly selected, it is impossible to infer your results upon a larger population of
maps” (Muehlenhaus 2011, p. 13). When the data is not sampled, as in this case, conclusions
cannot necessarily be applied to the whole area. This study’s method of data collection does not
sample data, but only documents the building use of a point that at one moment in time was a
cultural institution.
4.4 Diving Deeper
To overcome the drawbacks of the original un-sampled and limited dataset, a historically
significant section of Long Island City was chosen for a more in-depth analysis. Hunter’s Point,
the area in the south of Figure 4, was chosen as a new, more-localized study area that had
undergone a diverse transformation in the first half of the 1900’s. Ideally, a more detailed
analysis would show that Hunter’s Point thrived in the earlier part of the 20
th
century, but
experienced a large amount of industrial growth and vacant lots once the Queensborough Bridge
was built and the LIRR expanded to Queens Plaza. Since the commuters no longer came to
Hunter’s Point on their way to and from work, the cultural institutions and shops did not benefit
from the same amount of patronage anymore. Rather than focus on cultural institutions, this new
method documented the building use for every point that was once a building in the 5 x 3 block
area using the same methodology described in Chapter 3.
38
The 1891 Atlas of New York map could not be used for this more detailed study as its
classifications and illustrations differed too much from the four Sanborn maps and could not be
reconciled for lack of detail. Furthermore, since this dataset was to be more granular, there were
a variety of footprints to be accounted for. Digitizing building footprints would certainly be an
intriguing, though time-consuming, approach. However, this study continued documenting
points so that it would, firstly, utilize the same methodology and, consequently, be compared to
the previous, limited dataset.
To overcome the challenges of changing footprints—such as a group of buildings that
become an industrial complex—all identifiable buildings were digitized as a single point in all
map years, regardless of footprint (Figure 16). Since the maps were georeferenced to modern-
day LIC, a group of points that covered a large area that eventually became one building or
complex would be assigned the same category as the area that they existed in on the later map.
Each and every point that was once a building in all four years was digitized and categorized in
every other year using the same process as described earlier. By documenting the buildings as
points in this way, the expansion and retraction of footprints were taken into account.
39
Figure 16 Diagram demonstrating how shifts were determined based on the footprint of the
latter year’s categorization. Parcels at the corner of 11
th
Street and 50
th
Ave shifted from
residential lots (D), one shop (S), and one garage (A) to an industrial complex, which is
represented in the shifts of the points from 1936 to 1950.
Using the same method for identifying shifts (as described in Chapter 3), each point was
assigned a rating and each rating was compared to the previous year’s so as to determine the type
of shift that occurred. This allowed for a similar spatial analysis to be conducted while
meticulously tracking each and every building and how its use transformed through the time
periods.
Due to the nature of the collection of the first dataset, cultural points would logically
occur the most, but this does not lend itself to a comprehensive analysis of the study area. For
instance, when digitizing the new study area, it became apparent that a new category needed to
be included. In addition to the original five building types (Cultural, Industrial, Residential,
Shop, and Vacant) a new categorization known as “Other” was added, which grouped together
subtypes that were previously unneeded (Table 9). It was not that these subtypes had been
40
disregarded in the previous methodology, but they were simply not observed within the smaller
dataset. Such new categories were captured under the “Subtype” field, but grouped under the
“Other” type to consolidate the data. Whereas the vacant type describes a building that was
specifically labeled as such on the Sanborn map, the absent subtype described a space that either
had no building, or no label to indicate a building.
Table 9 Subtypes included under the Other type in the new dataset
Type Subtype
Other
Garage
Storage
Absent
Livestock
Transportation
4.5 The Story at the Local Scale
By not concentrating on cultural institutions and examining every building, it was now
possible to focus on all shift types(Table 10). In analyzing this data, however, it is important to
acknowledge that not every single point represents a building location in every year. These
points, as described in Figure 16, keep track of changing footprints, but do not denote the total
amount of building use types. Nevertheless, in studying the results, there are clear trends in shifts
of building use types.
Table 10 Number of each type of shift in Hunter’s Point
Type 1898 1915 1936 1950
Cultural 9 13 14 13
Industrial 27 47 114 141
Other 122 109 77 107
Residential 301 269 254 205
Shop 158 176 153 144
Vacant 1 4 6 7
41
Continuing by documenting the percentage increase from year to year, many more
fluctuations in point type became apparent (Table 11). Ultimately, within this smaller study area,
it was apparent that the cultural institutions did not, in fact, alter significantly in comparison to
many of the other building types. While there was a decent increase of cultural points between
1898 and 1915, the other time periods showed negligible changes. Other categories, such as
industrial and residential, had substantial variations from year to year. From 1898 to 1915 and
1915 to 1936, industrial points grew by 74.07% and 142.55%, respectively. Conversely,
residential points steadily decreased, losing 31.89% of the points between 1898 and 1936.
Table 11 Percentage change in point types between each time period
Type 1898 to 1915 1915 to 1936 1936 to 1950
Cultural +44.44% +7.69% -7.14%
Industrial +74.07% `+142.55% +23.68%
Other -10.66% -29.36% +39.96%
Residential -10.63% -5.58% -19.29%
Shop +11.39% -13.07% -5.88%
Vacant +300% +50% +16.67%
While these percentages are telling for the time, it is important to consider the difference
in range of years for each time period. While there are 17 years in the first time period, there are
19 in the next, and 14 in the final. The differences in ranges could explain why the shifts are
greater from 1915 to 1936. In order to take these variances into account, the following formula
calculated the relative rate of shifts per year in that time period:
42
where the Shift Rate identifies the increase or decrease of the type of shift(S) between two map
years (Y). Such a formula provides more context for the range of years and puts certain types of
shifts into perspective(Table 12).
Table 12 Shift rate in shifts per year
Type 1898 to 1915 1915 to 1936 1936 to 1950
Cultural +.24 +.05 -.07
Industrial +1.18 `+3.53 +1.93
Other -.76 -1.68 +2.14
Residential -1.88 -.79 -3.50
Shop +1.06 -1.21 -.64
Vacant +.18 +.11 +.07
While this interpretation of rate does not necessarily indicate what changes occurred from
year to year, it does show the increases and decreases as it relates to the map year differences and
total number of type variations. For instance, vacant shifts are given less importance due their
overall low count and their overall rate change is minimal. However, it is still evident that
industrial shifts show a dramatic increase throughout time. What is more, industrial shifts
increase at the same rate from 1915 to 1936 as residential shifts decrease from 1936 to 1950.
These two are the largest rates of change, which was not as evident in the percentage change.
Just as Figure 15 illustrates the shift of cultural institutions, Figure 17 illustrates the shifts
from 1898 to 1950 for all building uses. Within these maps, shifts, in general, seem to peak from
1915 to 1936, and by 1950, there are a large number of industrial locations. It is not only more
apparent in this denser dataset that Hunter’s Point experienced steady industrialization, but it is
more accurate to conclude that this was the case. Every point is accounted for, and therefore, the
entire study area, and not just a sample, is considered.
43
Figure 17 Building use shifts in Hunter’s Point from 1898 to 1950
44
As mentioned, the Hunter’s Point study area was chosen based on its historical relevance
within LIC. With this clearer and more diverse dataset, it was possible to compare the results to
the historical narrative developed in Chapter 2. This study’s principal reference for historical
LIC, “300 Years of Long Island City” by Vincent Seyfried, declares that Hunter’s Point became,
“a ghost town of shabby and neglected buildings” once the Queensborough bridge and Steinway
tunnels were opened (Seyfried 1984 p. 139). The sharp increase in industrial points and decrease
in residential points demonstrates the decrease in commuter patronage.
However, the other types of shifts do not speak to Seyfried’s assertion that Hunter’s Point
became a ghost town. In fact, despite increasing over time, vacant points only made up 1% of the
total points in 1950. The results ultimately challenge this portion of historical narrative.
In her essay, City as Space, City as Place: Sources and the Urban Historian, Carla
Pascoe attributes historians’ various views of the past to the types of sources they use(Pascoe
2010). She argues that oral history is the primary and most-reliable source, while urban planning
documents tend to be more skewed, since they were specifically created to solve a problem. In
forming his chronicle of Long Island City, Seyfried used a variety of historical references in
order to form his own mental map of the area. This mental map could be influenced by the types
of sources he used, forming a different representation of the area than that which had occurred.
Although it is well known that Hunter’s Point experienced decay during industrialization, it is
apparent that it was not necessarily a “ghost town”, since several cultural institutions persisted
and many shops still lined the streets. The results help to illuminate this remaining activity in the
sub-neighborhood, showing an area that could not solely be defined by industry and vacant
buildings.
45
4.6 Looking Backwards
The study of history is useful in documenting causes and effects throughout time
(Bodenhamer 2008). So far, however, this study has only documented what the locations have
shifted into, but not what they have shifted from. Determining the initial location type is
important so as to identify any transformation triggers. By rearranging the Python script in the
original ModelBuilder model, it was possible to track changes in the point based on its initial
category. Just as before, the script used an If-Then statement to compare the Shift Rating (sum) to the Rating for that particular feature. It determined both to and from categories by assessing
the difference between the sum and the current feature’s Rating. For instance, if the sum was 5,
and the current feature’s rating was 1 (industrial), then the point had shifted from a residential
location(Rating = 4) to an industrial location. A copy of this script is provided in Appendix B.
Table 13 shows the totals that the ModelBuilder model produced. Only transformations that had
at least one entry in one year appear, so if a specific type of transformation is not listed below
(such as Vacant to Shop) it did not occur.
As expected, many locations simply remained the same type of building, as this requires
the least amount of investment. However, “Other to Industrial” was the biggest type of shift.
Since the “Other” type consists of several subtypes, Table 14 shows how these “Other to
Industrial” shifts were split. Locations that shifted from absent (no building) to industrial were
the most numerous in this category. As the “Absent” subtype describes a nonexistent plot that
was not labeled “Vacant”, the data demonstrates that many of the industrial shifts occurred on
land that had not yet been developed. Since the original dataset did not capture the “Absent”
subtype, this could explain why there were not as many industrial shifts as expected. Many
industrial areas were not built on existing plots, but were built anew, which is supported by
46
Seyfried’s explanation that LIC was seen as a great area to install factories due to
underdevelopment of the land and proximity to Manhattan (Seyfried 1984).
Table 13 Initial building type and consequent building type through three time periods
Shift Type 1898 to 1915 1915 to 1936 1936 to 1950
Cultural no change 3 12 13
Cultural to Industrial 2
Cultural to Other 1
Cultural to Residential 1
Cultural to Shop 3
Cultural to Vacant 1
Industrial no change 14 38 93
Industrial to Other 5 6 15
Industrial to Residential 3
Industrial to Shop 4 2 6
Industrial to Vacant 1 1
Other no change 72 36 60
Other to Cultural 2
Other to Industrial 21 48 13
Other to Residential 8 10
Other to Shop 18 10 2
Other to Vacant 1 5 1
Residential no change 252 231 200
Residential to Cultural 6 1
Residential to Industrial 7 8 21
Residential to Other 19 19 26
Residential to Shop 16 10 2
Residential to Vacant 1 5
Shop no change 135 131 134
Shop to Cultural 2 1
Shop to Industrial 2 19 10
Shop to Other 13 13 4
Shop to Residential 6 12 5
Vacant no change 1
Vacant to Industrial 1 1 4
Vacant to Other 3 1
47
Table 14 The subtypes “Other” type and its shifts to industrial locations
Type 1898 to 1915 1915 to 1936 1936 to 1950
Absent to Industrial 20 42 11
Garage to Industrial 3
Livestock to Industrial 1
Storage to Industrial 4
Another predominant transformation occurred from residential to industrial building
types. As previously seen in Table 10, where residential points decreased from 301 to 205, these
latter results also show that residential buildings decreased over time. However, it is worth
recalling that due to the reuse of points during the data collection, each point does not necessarily
equate to a single building. It is likely that, as seen in Figure 13, several residential homes were
taken over by a giant industrial complex. Conversely, a large, singular building footprint could
turn into several, separate building points. Consequently, this data is telling, but does still not
represent the full story.
4.7 Identifying Other Types of Change
Documenting every single location allows one to look at the complete distribution of
individual categories and their shifts. For example, Figure 18 shows the distribution of industrial
shifts versus that of residential disappearances through all three time periods, which is an aspect
that the historical narrative did not address, and likely could not address, without GIS. The ovals
in the figure are standard deviational ellipses which summarize the general distribution and trend
of the points of each type.
48
Figure 18 Industrial Shifts and Residential Disappearances and their distributions over
three time periods
49
There does not appear to be a significant relationship when comparing the two
distributions, however, it is interesting to look at the results separately in relation to history. In
the 1898 to 1915 era, industrial shifts appeared to center in the northwest corner of the
neighborhood. This is likely due to the fact that the southeast corner contained the main
downtown area and had not yet felt the same effects of the Queensborough Bridge and Steinway
Tunnel construction.
In the following time period, 1915 to 1936, the industrial shift distribution seems to
disperse, while the residential disappearances align to the northwest. Since the ferries on the
western border had closed in 1925, it was likely that many inhabitants were no longer able to easily
commute to Manhattan so the residences were given up.
Finally, the 1936 to 1950 time period shows the residential disappearances tilting
southeast. While many residents had already moved out of the northwest, residents in the
southeast were now moving out. The benefit of such visualizations demonstrate the timing of
these effects. While the overall results challenge Seyfried’s assertion that Hunter’s Point became
a ghost town, these maps provide evidence to show that the sub-neighborhood still experienced
great industrialization and residential decline. Furthermore, the data and maps can help to show
where and when it occurred. A similarly in-depth study of other parts of LIC would surely reveal
more stories, specifically in Queens Plaza, which felt many effects contrasting to those of
Hunter’s Point.
50
CHAPTER 5: DISCUSSION AND CONCLUSIONS
This study tracked building use in LIC, New York in order to understand the neighborhood’s
industrial development. The category shifts in a point from one time period to the next
demonstrated what type of changes had occurred at a specific location. In documenting the
category of the location in both the preceding year and following year, it was possible to
determine what kind of changes took place in the neighborhood. These results can support,
disprove, and expand upon the historical narrative.
5.1 Implications and Limitations
In the first approach—using cultural points as a means of managing the magnitude of the
data collection—there were signs of industrialization as the number of industrial shifts picked up
from 1915 to 1950. However, the hypothesis was disproved, as many of the cultural institutions
persisted throughout time. In fact, there were five cultural institutions that persisted from 1891 to
1950, and seven that persisted through three out of four consecutive map years. This means that
these cultural locations did not experience any shift, whether it be industrial or another type.
Such lack of change demonstrates that cultural institutions were not significantly affected by
industrialization. Additionally, since only cultural locations were collected from year to year, the
dataset appeared limited. It was a singular story about the development of the cultural
institutions, but it did not give enough insight into the rest of LIC’s history.
The second approach, completed at a more localized scale, tracked the building use
category for all locations. Ultimately, it was concluded that to get more accurate results, it is
necessary to track every building through time. While this is more time consuming to collect, it
produces a complete documentation of the area and how it has developed. In this case study,
Hunter’s Point displayed more telling results. It showed that there were over five times more
51
industrial locations from 1898 to 1915. At the same time, it challenged the historical narrative,
showing that while Hunter’s Point experienced industrialization, the sub-neighborhood did not
become desolate.
The results also demonstrated a significant change in residential locations. Although
historical census information is not available at the neighborhood or parcel scale for LIC, the
data reveals a sharp decrease in residential points. The historical narrative addressed the decrease
of population in LIC as residents moved north to Astoria and this data shows exactly where and
when residential locations disappeared (Figure 18).
While this methodology cannot prove a connection between two events, it can help to
support it. In his own historical narrative of LIC, Vincent Seyfried attests that Hunter’s Point
declined due to the opening of the Steinway tunnels, which took residents away from the once
popular sub-neighborhood. The Vernon-Jackson station opened in 1915, and consequently
connected up to Queens Plaza, moving many residents out of Hunter’s Point. The 1936 map
helps support the cause and effect of this event by illustrating the decrease in residential points.
A good way of continuing to explore the cause and effect of this particular event would be to do
the same study of Queens Plaza, which will be discussed in the following section.
Nevertheless, there were still issues with digitizing all building points. For instance, it
was difficult to digitize every single point within every single year, while taking into account
footprint change. It was handled by setting categories for every point in every year within that
footprint, but this meant that there could be multiple points to represent one building.
Consequently, the tallies for all categories in these time periods do not accurately represent the
total number of the building types. In a way, the total numbers represent the total size of each
category, but even this cannot be precisely concluded.
52
In addition, by digitizing all points, a very detailed map is required. For historical maps,
this can be particularly difficult to find. In the case of this study, the 1891 Atlas of New York
map had to be passed over during the in-depth analysis, as it did not have the same detail as the
other Sanborn maps. To do such an analysis, one would need to locate historical maps that
documented the type of building use for every parcel and/or building.
5.2 Further Historical Conclusions
Creating a historical narrative is not only a foundation for the design of the data
collection process, it also offers a comparison for the results so that new historical insights can
be uncovered. Vincent Seyfried claimed that Hunter’s Point had become completely unoccupied
after the Queensborough Bridge and Steinway tunnels were built, but the data suggested that this
was not entirely true. Instead, these results—attained through the use of GIS—offer other stories
that the historical narrative had not considered.
In Figure 17, there are a few trends that are noticeable. For instance, in the southeast
block of Hunter’s Point, there is an area that experienced significant change. While it was
primarily residential in 1898, many of the points changed to the “Other” type in this area.
Looking at the Sanborn maps for 1915 and 1936, it appears that the parcels were wiped out, and
then in the 1950 map, the Midtown Highway appears, producing the “Transportation” subtype
points under the “Other” type. This structure was not documented in the historical narrative, but
in seeing this data, it is possible to go back and find the information that the narrative originally
overlooked. The Midtown Tunnel was built in 1940 to provide another passage for automobiles
to go to/from Manhattan and Queens(MTA 2015). The tunnel emerged at Vernon Boulevard and
Borden Avenue and continued as a highway over this portion of Hunter’s Point. The Midtown
Highway did not necessarily cause the disappearance of the residential buildings, but the results
53
help to show changes in Hunter’s Point that were not previously considered. Furthermore, it may
explain why the area surrounding the highway experienced industrial and absent shifts.
Some of the results were simply out of scope of the historical narrative, but now pose
new and intriguing questions about the area at the time. For instance, it appears that the
northwest portion of Hunter’s Point experienced the greatest amount of industrialization. While
it had a large amount of residential and shop locations in 1898, it was full of industrial points in
1950. If Vincent Seyfried had seen this data, he might have questioned why Hunter’s Point did
not appear to be a ghost town, as he had contended. Nevertheless, this northwest portion seemed
to have experienced some of the effects he suggested. This might have caused Seyfried to
wonder why this specific area of Hunter’s Point became particularly industrialized, while the rest
remained active.
Another interesting observation is the shopping corridor that runs north to south on
Vernon Boulevard. Locations change building type throughout Hunter’s Point, but this street in
particular seems to alter very little. Instead, the shops persist throughout all other changes. This
is another aspect of LIC’s history that was not addressed in the historical narrative, but would
nevertheless be interesting from a historian’s perspective. One might ask why these shops
persisted when and where they did. Perhaps there were zoning laws, or perhaps these stores
retained adequate patronage sufficient enough to keep them in business. These are aspects of
LIC’s history, discovered through GIS, which help form new stories of the area.
5.3 Future Work
After exploring the various methods for documenting change using historical maps, it is
evident how this method can be extended to other regions and studies.
54
5.3.1 Applying Methods Elsewhere
As the Hunter’s Point analysis proved effective in tracking change, it would be beneficial
to use the same approach for the entire area of LIC. In documenting every building use location,
it would be possible to see how the neighborhood changed and whether there were any particular
patterns that could be identified. Both the cultural points and historical narrative demonstrated
that Hunter’s Point and Queens Plaza were, at different times, transportation hubs. There was an
inverse relationship between these two areas, for once Queens Plaza was built, Hunter’s Point
declined. Such relationships would be interesting and beneficial to visualize, especially in
relation to the rest of the study area.
Including Astoria in the study area would also be valuable. This neighborhood in many
ways, was the antithesis of LIC. Although it did experience a brief stint of industry in the late
19
th
century, it eventually turned into a residential area. Just as Hunter’s Point experienced
residential decline, one would expect Astoria to experience a drastic increase in new homes.
These type of historical events and relationships could be explored if the same process was
carried out on these larger study areas.
Ideally, this study could also be completed in other areas, even those that do not have the
same industrial history. While LIC’s history in particular is industrial, this process can be used to
track land use of any type of neighborhood with any types of categorizations. The main benefit
to this method is the way in which it collects and organizes data from historical maps. In
managing building use as points from a historical map, one can analyze shifts from year to year.
These shifts can have a variety of categorizations that are specific to both the neighborhood and
the time, making it useful for other study areas.
55
5.3.2 Expanding the Research
Generally, an important aspect of land use is footprint. Whereas this project deliberately
focused on points to emphasize existence rather than size, another study could develop a similar
process that not only looks at shifts in building usage, but shifts in the total area, and even
volume of the building. For instance, Figure 12 did this for the large Ravenswood Park that
disappeared from 1891 to 1898. The results are certainly noticeable and impressive as factories
took over the area. Such results provide other insights into how the land was used and how a
single building developed spatially over time. The negative aspect of this approach, however, is
that the manual process cannot be streamlined in the same way. It would be necessary to turn to
automated line and feature recognition tools to move this to a much larger area. Furthermore, the
buildings cannot be compared easily, as footprints will move and shift over the landscape. Using
points allows one to compare a singular location from one period to the next, despite its
magnitude.
In Lehigh University’s digital library project, Beyond Steel, a group documented
Bethlehem Steel’s employees within Bethlehem city. The project not only used Sanborn maps,
but also books, photographs, and oral histories to get a complete spatial representation of the
employees, as well as a detailed database organizing their names, jobs and spouses. Building a
database such as this is time-consuming, yet the result is highly informative. It appeals to both
GIS analysts and historians alike in that it gathers both quantitative data and historical details
unique to the person and place. While outside of the scope of this research, forming a similarly
detailed database would be a great continuation. Like Beyond Steel, it would be informative to
see the location and movement of the LIC factory workers. This thesis demonstrated a movement
56
of residents, conjecturing that the workers of the factories moved to Astoria to live in a more
residential and less industrial area, but tracking this explicitly would be insightful.
5.4 Summary
This thesis ultimately created a process for creating a spatiotemporal database from
historical fire-insurance maps. While the approach to data collection changed from cultural
institutions to all building points in order to gain a greater insight into land use change, the
overall process provides a way of digitizing, organizing and analyzing data within historical
maps. The results helped to document LIC’s industrialization and its effect on other aspects of
the neighborhood as it related to the historical narrative. Ultimately, this method can be
expanded upon and used within other study areas under different search criteria to gain a greater
understanding of the land and time.
57
REFERENCES
Asadorian, W. & Seyfried, V. 1991. Old Queens, N.Y., in Early Photographs, ed. A. Weissman.
Mineola, NY: Dover Publications, Inc.
Bodenhamer, D. J., ed. 2010. The Spatial Humanities. Bloomington, IN: Indiana University
Press.
Brooklyn/Queens Waterfront. Ravenswood. 2012. Accessed April 4, 2015. https://sites-google-
com.libproxy.usc.edu/site/brooklynqueenswaterfront/neighborhood-histories/ravenswood
Buiso, G. 2013. “’Island’ nabe: Call Us LIC!” New York Post 2013.
Bytes of the Big Apple. 2014. 2010 Census Tracts & Blocks. City of New York: NYC
Department of City Planning.
Gilliland, J., and S. Olson. 2010. Residential segregation in the industrializing city: A closer
look. Urban Geography 31 (1): 29-58.
Greenberg, K. A brief history of LIC. 2008. Available from http://www.licweb.com/history.html.
Gregory, I. N. 2008. "A map is just a bad graph: Why spatial statistics are important in historical
GIS”, In Placing History, edited by A.K. Knowles and A. Hillier, 123. Redlands, CA: Esri
Press.
History of Long Island City. 2011. Queens West Villager 2011.
Kelsey, J. S. 1896. History of long island city, new york. New York, NY: Long Island Star
Publishing Company.
Kemp, K. K. 2009. What can GIS offer history? Journal of Humanities & Arts Computing: A
Journal of Digital Humanities 3 (1/2): 15-19.
Knowles, A. K. and Hillier, A., ed. 2008. Placing History. 1st ed. Redlands, California: ESRI
Press.
Lynch, K. A. 1960. The image of the city. The MIT Press.
Mahdavi, A., and H. Azizmohammadlou. 2013. The effects of industrialization on social capital:
The case of Iran. International Journal of Social Economics 40 (9): 777-796.
Maserjian, J. et al. 2014. Beyond Steel. Lehigh University. Accessed October 22, 2014
http://digital.lib.lehigh.edu/beyondsteel/
58
Massey, D. 2013. Doreen Massey on space. Social Science Bites. Transcript of podcast.
Accessed April 11, 2015 http://www.socialsciencespace.com/2013/02/podcastdoreen-
massey-on-space/
Mohan, G., & Mohan, J. 2002. Placing social capital. Progress in Human Geography,
(26)2, 191–210.
Montello D. R. and Sutton P. C. 2006. An Introduction to Scientific Research Methods in
Geography. Thousand Oaks, California: Sage Publications, Inc.
MTA. 2015 Queens midtown tunnel. Accessed May 14, 2015
http://web.mta.info/bandt/html/qmt.html
Muehlenhaus, I. 2011. Another goode method: How to use quantitative content analysis to study
variation in thematic map design. Cartographic Perspectives 69: 7.
NYC Planning. 1993. Plan for Long Island City: a framework for development. Accessed May
14, 2015 http://www.nyc.gov/html/dcp/html/pub/lic.shtml
Old Long Island Mansions. 1894. New York Times, March 18, 1894.
Oulton, A. 2012. Community gardens for social capital: a site suitability analysis in Akron, Ohio.
MS thesis, Spatial Sciences Institute, University of Southern California.
Pascoe, C. 2010. City as space, city as place: Sources and the urban historian. History Australia
7 (2): 30.
Pratt, E. E. 1911. Industrial causes of congestion of population in New York City. New York:
Columbia University.
Putnam, R. D. 1995. Bowling alone: America’s declining social capital. Journal of
Democracy, 6(1), 65-78.
Raymond, A. 2011. Denny regrade, 1893-2008: A case study in historical GIS. Social Science
History 35 (4): 571-597.
Ristow, W. W. 2014 Introduction to the Sanborn map collection. In Geography & Map Reading
Room. Accessed March 1, 2015 http://www.loc.gov/collections/sanborn-maps/articles-and-
essays/introduction-to-the-collection/
Sanborn. 1898. Borough of Queens City of NY, volume one. sheets 2-32, 63, 71-83.
Sanborn. 1915. Borough of Queens City of NY, volume one. sheets 2-32, 63, 71-83.
Sanborn. 1936. Borough of Queens City of NY, volume one. sheets 2-32, 63, 71-83.
59
Sanborn. 1950. Borough of Queens City of NY, volume one. sheets 2-32, 63, 71-83.
Seyfried, V. F. 1984. 300 years of Long Island City 1630-1930. New York: Queens Historical
Society.
Stadler, D. 2014. The History of Long Island City. Accessed October 10, 2014.
https://derekstadler.wordpress.com/portfolio-2/collections-photography/maspeth-from-
small-dutch-community-to-part-of-the-great-metropolis-final/the-history-of-long-island-
city-details-of-its-short-lived-days-as-both-an-incorporated-municipality-and-the-major-
western-terminus-of-the-long-island-rail-road/.
Hapke, C. G. and Thieler, E. R. 2011. USGS science for the nation's changing coasts: Shoreline
change research. USGS Fact Sheet. 2 p.
Velsey, K. 2013. Long Island City is having an identity crisis. New York Observer 2013.
Vitullo-Martin, J. 2004. Rezoning Hunters Point. Center for Rethinking Development, The
Manhattan Institute. Accessed November 6, 2014 http://www.manhattan-
institute.org/email/crd_newsletter07-04.html
Wolverton, C. 1891. Part of Long Island City, Ward No. 1 & 3. Atlas of Queens Co., Long
Island, New York.
60
APPENDIX A: Georeferenced Maps
Figure 19 The Atlas of New York 1891 Long Island City map georeferenced to
modern-day LIC
61
Figure 20 Sanborn 1898 Long Island City map georeferenced to modern-day LIC
62
Figure 21 Sanborn 1915 Long Island City map georeferenced to modern-day LIC
63
Figure 22 Sanborn 1936 Long Island City map georeferenced to modern-day LIC
64
Figure 23 Sanborn 1950 Long Island City map georeferenced to modern-day LIC
65
APPENDIX B: Model Builder Model
Figure 24 Model Builder model that organizes cultural institution point data.
66
Figure 25 Model Builder model to determine type of shifts in between time periods
67
Figure 26 Python script assigning shift types based on previous year's category
Abstract (if available)
Abstract
The goal of this thesis was to develop a process in which historical land use can be tracked in order to gain a better understanding of an area’s history. The study area, Long Island City (LIC) is historically an industrial neighborhood within Queens County of New York City. By documenting its land use shifts from 1891 to 1950, it is possible to visualize and analyze the changes that occurred as industrialization took place. ❧ This study compiles a digital historical narrative to provide a foundation for understanding the data, as well as a reference for making new conclusions from the results of the analysis. Old fire insurance maps provide building footprints categorized by use. These were used to digitize locations of interest as points that were catalogued under five different categories: Cultural, Industrial, Residential, Shop, and Vacant at each of five time periods. The resulting spatiotemporal database makes it possible to track a single building and its use through a period of 59 years. ❧ The methodology developed for this thesis collects and classifies building use as points so as to develop efficiently and quickly an accurate historical dataset. In doing so, the project tracked the cultural development of LIC through an examination of a set of key buildings, as well as the overall land use change of a sub-neighborhood, Hunter’s Point. It determined that by tracking the use of every building through every map year, one gets a better historical analysis. Such methods can be used not only to help support previously known historical narratives, but also to allow for new conclusions to be drawn.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Development of a historical urban land use web application for the city of Hong Kong
PDF
Using pattern oriented modeling to design and validate spatial models: a case study in agent-based modeling
PDF
Delimiting the postmodern urban center: an analysis of urban amenity clusters in Los Angeles
PDF
Investigating electoral college reform: geography's impact on elections, and how maps influence our perception of election outcomes
PDF
Out-of-school suspensions by home neighborhood: a spatial analysis of student suspensions in the San Bernardino City Unified School District
PDF
Validation of volunteered geographic information quality components for incidents of law enforcement use of force
PDF
Exploring land use changes in the city of Irvine’s master plan
PDF
The geography of voter power in the U.S. electoral college from 1900-2012
PDF
Walking to the Longhouse: a deep map of the Central New York Military Tract & its indigenous history
PDF
The role of GIS in asset management: integration at the Otay Water District
PDF
Spatiotemporal visualization and analysis as a policy support tool: a case study of the economic geography of tobacco farming in the Philippines
PDF
A fire insurance map geocoder for pre-earthquake San Francisco
PDF
Exploring the impact of the natural gas leak on home values in Porter Ranch, California
PDF
Automating “Ethington Transections”: a new visualization tool
PDF
Generating bicyclist counts using volunteered and professional geographic information through a mobile application
PDF
A site suitability analysis for an inland port to service the ports of Los Angeles and Long Beach
PDF
Residential housing code violation prediction: a study in Victorville, CA using geographically weighted logistic regression
PDF
Surface representations of rainfall at small extents: a study of rainfall mapping based on volunteered geographic information in Kona, Hawaii
PDF
Selection of bridge location over the Merrimack River in southern New Hampshire: a comparison of site suitability assessments
PDF
Social media to locate urban displacement: assessing the risk of displacement using volunteered geographic information in the city of Los Angeles
Asset Metadata
Creator
Mamer, Elizabeth J.
(author)
Core Title
Exploring urban change using historical maps: the industrialization of Long Island City (LIC), New York
School
College of Letters, Arts and Sciences
Degree
Master of Science
Degree Program
Geographic Information Science and Technology
Publication Date
06/22/2015
Defense Date
05/08/2015
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
GIS,HGIS,History,LIC,Long Island City,New York City,NYC,OAI-PMH Harvest,Urban
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Kemp, Karen K. (
committee chair
), Ethington, Philip J. (
committee member
), Vos, Robert O. (
committee member
)
Creator Email
ejmamer@gmail.com,emamer@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-576264
Unique identifier
UC11299520
Identifier
etd-MamerEliza-3506.pdf (filename),usctheses-c3-576264 (legacy record id)
Legacy Identifier
etd-MamerEliza-3506.pdf
Dmrecord
576264
Document Type
Thesis
Format
application/pdf (imt)
Rights
Mamer, Elizabeth J.
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
Tags
GIS
HGIS
LIC
NYC