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
/
Development of a historical urban land use web application for the city of Hong Kong
(USC Thesis Other)
Development of a historical urban land use web application for the city of Hong Kong
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
DEVELOPMENT OF A HISTORICAL URBAN LAND USE WEB APPLICATION FOR THE
CITY OF HONG KONG
by
Brandon N. Patton
A Thesis Presented to the
FACULTY OF THE USC DORNSIFE COLLEGE OF LETTERS, ARTS AND SCIENCES
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(GEOGRAPHIC INFORMATION SCIENCE AND TECHNOLOGY)
August 2023
Copyright 2023 Brandon N. Patton
ii
To my family, for their continuous support
iii
Acknowledgements
I would like to thank Dr. Vos and Dr. Wilson, for their insights and guidance in my geospatial
journey. I am especially grateful for Dr. Sedano’s patience and help clarifying the direction of
this project. I am grateful for the works of Vivian Ng and Tymon Mellor that inspired this thesis.
Lastly, I would like to thank Vito Chiu from the Hong Kong Government Records Office for his
help securing the source maps used in this project.
iv
Table of Contents
Dedication ...................................................................................................................................... iii
Acknowledgements ........................................................................................................................ iv
List of Tables .................................................................................................................................. vi
List of Figures ............................................................................................................................... vii
Abbreviations ................................................................................................................................. ix
Abstract ............................................................................................................................................ x
Chapter 1 Introduction ..................................................................................................................... 1
1.1 Hong Kong Background ...................................................................................................... 1
1.1.1 Colonial History of Hong Kong ................................................................................. 2
1.1.2 Hong Kong Geography ............................................................................................... 3
1.1.3 Hong Kong Economic Growth and Land Reclamation .............................................. 6
1.2 Land Use and Urban Development in Hong Kong ............................................................. 7
1.2.1 Urban Planning for Density ........................................................................................ 7
1.2.2 Planning Department in HK ....................................................................................... 8
1.3 Open Data Overview ......................................................................................................... 10
1.4 Application Overview ....................................................................................................... 10
1.4.1 Target Users .............................................................................................................. 11
1.4.2 Study Area ................................................................................................................ 11
1.4.3 Data Overview .......................................................................................................... 12
1.4.4 Data Accessibility ..................................................................................................... 13
1.4.5 Methodology Overview ............................................................................................ 13
1.5 Thesis Structure ................................................................................................................. 14
Chapter 2 Related Work ................................................................................................................ 15
2.1 Historical GIS (HGIS) ....................................................................................................... 15
2.1.1 Importance of HGIS in Scholarship ......................................................................... 15
2.1.2 Examples of HGIS Research .................................................................................... 18
2.1.3 Examples of HK HGIS ............................................................................................. 21
2.2 Open Data and Open Source GIS ...................................................................................... 22
2.2.1 What is Open Data? .................................................................................................. 23
2.2.2 Origins of Open Data ................................................................................................ 24
2.2.3 Open GIS .................................................................................................................. 24
2.2.4 Open Geospatial Data ............................................................................................... 26
2.2.5 Opportunities and Challenges ................................................................................... 28
2.2.6 Urban and Land Use Open Data/GIS ....................................................................... 30
Chapter 3 Methodology ................................................................................................................. 32
3.1 Requirements and Objectives ............................................................................................ 32
v
3.1.1 Accessibility ............................................................................................................. 32
3.1.2 Functionality and Intended Users ............................................................................. 33
3.1.3 Software Requirements ............................................................................................ 34
3.2 Software Choice ................................................................................................................ 34
3.3 Data .................................................................................................................................... 35
3.3.1 Map of TPU 114 (1972) ........................................................................................... 36
3.3.2 Map of TPU 114 (1982) ........................................................................................... 37
3.3.3 Map of TPU 122 (1967) ........................................................................................... 38
3.3.4 Map of TPU 122 (1985) ........................................................................................... 39
3.3.5 Map of TPU 123 (1967) ........................................................................................... 40
3.3.6 Map of TPU 123 (1985) ........................................................................................... 41
3.4 Building the Historical Database ....................................................................................... 43
3.4.1 Georeferencing ......................................................................................................... 43
3.4.2 Digitizing Historical Maps/Creation of Spatial Data ............................................... 45
3.5 Web App Development ..................................................................................................... 45
3.5.1 Publishing Web Layers ............................................................................................. 46
3.5.2 Designing the Web Map ........................................................................................... 46
3.5.3 Designing the Web App ........................................................................................... 49
3.5.4 Publishing the Webapp ............................................................................................. 52
Chapter 4 Results ........................................................................................................................... 53
4.1 Root Mean Square of Control Points ................................................................................. 53
4.2 Spatial Data Created .......................................................................................................... 54
4.3 Web Map ........................................................................................................................... 60
4.4 Web App ............................................................................................................................ 61
Chapter 5 Discussion ..................................................................................................................... 71
5.1 Project Successes ............................................................................................................... 71
5.2 Utility of the Web App ...................................................................................................... 72
5.3 Esri Versus Open Geospatial Software ............................................................................. 74
5.4 Challenges/Limitations ...................................................................................................... 75
5.5 Future Work ....................................................................................................................... 76
References ..................................................................................................................................... 78
vi
List of Tables
Table 1. Land use survey map data ............................................................................................... 36
Table 2. RGB values for land use categories ................................................................................ 48
Table 3. RMS Errors Table ........................................................................................................... 54
vii
List of Figures
Figure 1. Map of Hong Kong SAR ................................................................................................. 4
Figure 2. Aerial imagery of Hong Kong Island ............................................................................... 5
Figure 3. Boundary of TPUs demark study area ........................................................................... 12
Figure 4. Original map of TPU 114, year 1972 ............................................................................. 37
Figure 5. Original map of TPU 114, year 1982 ............................................................................. 38
Figure 6. Original map of TPU 122, year 1967 ............................................................................. 39
Figure 7. Original map of TPU 122, year 1985 ............................................................................. 40
Figure 8. Original map of TPU 123, year 1967 ............................................................................. 41
Figure 9. Original map of TPU 123, year 1985 ............................................................................. 42
Figure 10. Map of TPU 123 layer from year 1967 ........................................................................ 55
Figure 11. Map of TPU 123 layer from year 1985 ........................................................................ 56
Figure 12. Map of TPU 122 layer from year 1967 ........................................................................ 57
Figure 13. Map of TPU 122 layer from year 1985 ........................................................................ 58
Figure 14. Map of TPU 114 layer from year 1972 ........................................................................ 59
Figure 15. Map of TPU 114 layer from year 1982 ........................................................................ 60
Figure 16. Web Map of all TPU layers in ArcGIS Online ............................................................ 61
Figure 17. Landing page of the web application ........................................................................... 62
Figure 18. Layout of web app with all widgets engaged ............................................................... 63
Figure 19. Layout of widgets open and moved across page .......................................................... 64
Figure 20. Example of Swiping widget ......................................................................................... 65
Figure 21. Attribute table opened from bottom of app screen ...................................................... 66
Figure 22. Filtering Business and Office land use type from layer TPU114 1972 ....................... 67
viii
Figure 23. Screening widget opened to access shapefile download function ............................... 68
Figure 24. About widget opened with introduction and user tips ................................................. 69
Figure 25. Add Data widget opened with original coastline and land reclamation data added .... 70
ix
Abbreviations
DTM Digital Terrain Model
GIS Geographic information system
GISci Geographic Information Science
GIST Geographic information science and technology
HK Hong Kong
HULU Historic urban land use
OZP Outline Zoning Plan
PlanD Planning Department
SAR Special Administrative Region
TPB Town Planning Board
TPU Tertiary Planning Unit
x
Abstract
Hong Kong is a dynamic city located on the southern coast of mainland China. A once
unassuming island of fishing villages became an economical trading hub as a British crown
colony following the Opium Wars. Hong Kong’s geology limited the natural area of developable
land, and as the population of the colony increased over the decades, land reclamation projects
were commissioned to account for exponential emigration. Over the course of its century-long
history, the urban topography of Hong Kong has transformed significantly, and is expected to
continue to evolve as the city maintains its status as an international business hub. This thesis
explores the development of an open Historical Geographic Information System (HGIS) web
application that portrays interactive digital versions of historic land use maps of Hong Kong.
Historic maps of Tertiary Planning Units in Hong Kong’s Central district from the late 1960s to
the late 1980s are the maps used in this application’s first iteration. This project incorporates
georeferencing and spatial data creation techniques and methodologies for digitizing historical
data and configuring app building software. The Hong Kong Historic Urban Land Use Web App
successfully delivers the historic data in a user-friendly web-based application that allows users
to investigate the land use differences and download the complete dataset for their individual
purposes. The project also serves as a model for the development of similar web-based
applications for exploring historical spatial data.
1
Chapter 1 Introduction
Hong Kong is a unique and dynamic city located near the Pearl River Delta off China’s southern
coast. The city is rich with history interwoven with its complex geography. Shaped by its
position as an important former British colony, the city has undergone periods of rapid
population growth and urbanization to become a global financial hub. A key feature of Hong
Kong’s geography is the land reclamation projects that have altered the original coastlines of
Hong Kong Island and the Kowloon peninsula. Consequently, land reclamation profoundly
impacted Hong Kong’s land use and urban development.
This thesis project creates an interactive web map application that shares spatial data of
historical land use and land zoning of Hong Kong’s Central and neighboring districts. Prior to
this, these maps were only available in print form and could only be accessed through designated
record offices. This application utilizes a web geographic information system (GIS) to provide
downloadable open datasets of these historical maps as well as allow users to search and query
the data based on land use, town planning unit, and date. This chapter provides more background
on Hong Kong’s history, land reclamation and its relation to the land use (as fifty-two percent of
the spatial data captured sits on reclaimed land), and an overview of the project goals, users, and
development process.
1.1 Hong Kong Background
The Hong Kong Historical Urban Land Use Web App provides historic Hong Kong GIS
data in a digital format. To better understand the context of this application, this section provides
background information on Hong Kong’s history.
2
1.1.1 Colonial History of Hong Kong
Originally a cluster of small fishing villages, Hong Kong eventually became an important
free port for traders of the British Empire. Early in the 19th century, the British Empire was
dependent on tea imports from China to meet its local demand. During this period, Imperial
China only accepted payments of silver. The British Empire could not sustain this exchange and
began to illegally import opium into China (Civatis n.d.).
The first Opium War started in 1839 when the Qing dynasty declared opposition to the
opium trade. After China’s defeat, the island of Hong Kong was ceded to the British under the
Treaty of Nanking in 1842. The Second Opium War followed, and it led to further Chinese
cessions. The 1860 Convention of Beijing granted the British control of the Kowloon Peninsula
and Victoria Harbor. In a final convention over territory in 1898, the full extent of British Hong
Kong also included the New Territories and Lantau Island, which were given a lease term of 99
years, set to expire on June 30, 1997 (Szcepanski 2020).
The Chinese Revolution in 1911 brought the establishment of the Republic of China in
1912, and Hong Kong became a political refuge for exiles, prompting an initial growth in
population. During World War II, Hong Kong briefly fell under Japanese control, but the British
took it back after the Japanese surrender in 1945. The Chinese Civil War brought an influx of
refugees from the mainland, both rich and poor. This base of labor and capital set the foundation
for Hong Kong’s growth towards an international manufacturing and financial hub (Schenk
2023). The decades following saw an increasing influx of immigrants from southern China and
fueled rapid economic and population growth. By the end of the 1980s, Hong Kong had become
one of the most prosperous places in Asia.
At the approach of the end of the British lease on the New Territories, China and Britain
signed a Joint Declaration on December 19, 1984, to pronounce Hong Kong a Special
3
Administrative Region (SAR) of China. The agreement of this declaration brought forth the,
“one country, two systems” policy, which would allow Hong Kong to retain its existing legal
system and capitalist economy for 50 years following the handover (Civatis n.d.). On July 1,
1997, Hong Kong officially became Hong Kong SAR, marking the historic handover.
1.1.2 Hong Kong Geography
The Hong Kong SAR is located at the southern coast of China at the mouth of the Pearl
River. The territory encompasses the two major islands of Hong Kong and Lantau, a portion of
China’s mainland (the Kowloon Peninsula and New Territories), as well as a collection of 230
smaller islands and the waterways that interconnect them. The total area of the territory is 2755
square kilometers, roughly 1104 square kilometers (40% of total area) of which is comprised of
land. Figure 1 shows the official territory of the Hong Kong SAR.
4
Figure 1. Map of Hong Kong SAR
The area’s topography is considered rugged, constituting steep mountains and deep
valleys. The highest point is Tai Mo Shan located in the New Territories with an elevation of
957m (CEDD 2019). Figure 2 shows modern aerial imagery of Hong Kong Island, the Kowloon
peninsula, and neighboring islands.
5
Figure 2. Aerial imagery of Hong Kong Island
The geology of Hong Kong is an important aspect of understanding the full picture of its
urban development. Most of Hong Kong is igneous rock, specifically granitic and volcanic rocks.
Igneous rock accounts for 85% of Hong Kong’s land surface while sedimentary rock accounts
for 15% (Shaw 2010). Igneous rock is extremely hard and dense, making it difficult to excavate
or drill through, they are considered unfavorable for engineering purposes (Blat 2006). Because
84% of the total area of Hong Kong is sloped (Ye 1998), developing and building on land poses
a unique issue. Hong Kong’s hills are generally slopped at 30-45 degrees (Gregory 1964), only
adding to the engineering difficulty.
6
1.1.3 Hong Kong Economic Growth and Land Reclamation
There is a high demand for land space in economic development. The coastal region of
China is one of the most densely populated places in the world and it is a main component of the
fast-growing national economy. In 2011, China’s coastal provinces and cities contributed to
60.8% of the national GDP, while they hosted 43.5% of the country’s population in only 13% of
its territory (Liu and Liu 2008). The combination of a large population and a rapid-growing
economy has led coastal administrations to promote urbanization and expand. China imposes a
minimum area of land that all regions should maintain for food production and the strictness of
this policy means that developing on arable land is not possible. In response to this, local
governments have turned to land reclamation as a replacement for the arable land that would
have been used for urban or industrial development (Liu and Liu 2008).
In contrast to other world cities, the developed area of Hong Kong is only 25% of the
entire territory, highlighting how much the region is hamstrung by its topography. Hong Kong
has undergone a series of land reclamation projects in an effort to accommodate the expanding
population and work around the area’s naturally difficult geology. Land reclamation is the
process of creating new land by filling in existing bodies of water. In total, the land reclamation
programs have added 70 square kilometers to the city’s area. The first land reclamation project
occurred in 1851 in the western part of Hong Kong Island after a fire destroyed over 400 homes
and the British rulers decided to combine the rubble and soil and deposit it in the harbor to create
a new roadway. Between then and the early part of the 20th century, most land reclamation
projects occurred on the northern coast of Hong Kong and were welcomed as a way to alleviate
health and safety concerns of overcrowded neighborhoods. As of 2018, the sections of reclaimed
land comprised about 6% of the total area of Hong Kong and 25% of its developed land (Ng
2022).
7
1.2 Land Use and Urban Development in Hong Kong
Land use is a vital component of urban planning because it determines how land is
allocated and developed in cities. Good land use planning ensures that the urban environment is
functional, efficient, and sustainable. Land use should also align to support the economic,
environmental, and social goals of the city. Because land in Hong Kong is an extremely scarce
resource, effective land use planning is paramount for managing the city’s growing populations
and the demands of its multifaceted economy.
The “Hong Kong 2030+” planning vision is a policy initiative the Hong Kong
Government has implemented to manage the city’s land resources and guide land use planning
and development (Planning Department 2021). In this plan, the government aims to promote
more sustainable land use practices such as development of green spaces and promotion of
developments with low-carbon footprints.
1.2.1 Urban Planning for Density
Cities like Hong Kong and Singapore successfully incorporated the high-rise, high
density Multiple Intensive Land Use (MILU) concept in the late 20th century (Zhang 2000).
MILU involves increasing the density of land use by combining residential, commercial, and
other activities in urban areas that are supported by public transportation infrastructure. Hong
Kong only has 17,600 hectares of buildable land to house 6.7 million people, resulting in a high-
density range of 2500-3000 people/ha (Lau 2005). The population of Hong Kong is projected to
increase steadily over the next 30 years with an average rate of 0.7% per annum (Wang et al.
2015). There is a basic housing demand for the growing population. The average living space in
Hong Kong is less than 14 square meters per capita (Wong 2011). Leading up to 2030, the total
8
housing demand is 924,000 units (Wang 2015). Additional land is also needed to serve economic
activities and transportation infrastructure.
Because Hong Kong is one of the most densely populated urban areas in the world,
focusing on vertical expansion has been the key solution to managing limited land and resources.
Urban designers and managers have shown great interest in Hong Kong’s approach to vertical
land use, recognizing its potential to maintain a thriving living and work culture (Lau 2005).
Developing effective land use policies requires consideration of local characteristics. In
cities with high population density, compact, mixed, and efficient land use patterns are preferred.
In contrast, in less dense cities, urban layouts that prioritize convenient living for citizens are
prioritized over purely efficient use of land (Wang et al. 2015). As a typical high-density city at
the crossroads of fast population growth and limited land resources, Hong Kong’s high-rise
approach was the logical choice. Sustainable development and land use is still a crucial part of
Hong Kong’s urban plan.
1.2.2 Planning Department in HK
There are two entities responsible for the urban planning and development of Hong
Kong, The Town Planning Board (TPB), and the Planning Department (PlanD). The Town
Planning Board is a statutory body of the Hong Kong Government that is tasked with developing
urban plans. Members are appointed by the Chief Executive including professionals from sectors
such as architecture, engineering, and surveying. It is established under the Town Planning
Ordinance with the main goal of promoting health, safety, convenience, and general welfare of
the community through the preparation of layout plans for areas of Hong Kong. Its functions
including considering planning applications, making amendments recommendations on
9
development plans, and listening to public’s representation on planning matters (Town Planning
Board n.d.).
The Planning Department and is the organization that decides how land in Hong Kong
can be used. It is responsible for implementing the land use planning policies and decisions set
by the TPB. The objective of the Planning Department is to foster a better live and work
environment for residents of Hong Kong. The Planning Department is responsible for the
preparation of different types of plans to guide the proper use and development of land. The
plans can range from developmental strategies at the territorial level to statutory and
departmental plans at the district (Planning Department 2018).
An Outline Zoning Plan (OZP) is a statutory plan prepared by the Town Planning Board
under the Town Planning Ordinance. Essentially it is a plan that portrays the land-use zonings
and major road systems of individual planning areas. Land within the maps are divided into
different zones marked with their land uses, such as commercial, industrial, or residential. A
zone can also be divided into subtypes; for example, residential could have several subtypes
depending on the proximity to the coastline. Each plan is accompanied by a Schedule of Notes
which specifies the uses have always been permitted and those that would require permission
from the Town Planning Board (Planning Department 2018).
The key relationship between the TPB and the Planning Department is that the TPB sets
the policy framework for planning decisions, while the Planning Department implements those
decisions and provides technical expertise.
10
1.3 Open Data Overview
The simple definition of open data is data that “can be freely used, re-used, and
redistributed by anyone – subject only, at most, to the requirement to attribute and sharealike”
(Open Data Handbook n.d.). The open data movement is also being used by governments to
release their data to make it available for the public to use. This open data can then be used by
citizens to develop new applications and services. It also increases governmental transparency
and promotes more accountability of responsible decision making.
Hong Kong promotes its open data initiative with its website data.gov.hk. Data.gov.hk is
coordinated by the Office of the Government Chief Information Officer with the participation of
other government departments to disseminate different types of information for the public sector.
The objective of the data.gov.hk portal is to increase community value by providing
demographic, economic, geographical, meteorological, and historical data to the public to be
used in commercial and non-commercial purposes (data.gov.hk n.d.)
1.4 Application Overview
The goal of the Hong Kong Historical Urban Land Use Web App is to provide an
accessible database where historic land use data of Hong Kong can be easily downloaded and
used for other spatial science work. Currently land use maps and records for Hong Kong are only
accessible by visiting the Government Records or Planning Department offices in person and
searching based on Town Planning Unit or Outline Zoning Plan. This greatly limits the work that
can be done surrounding historic land use in Hong Kong. The main utility of this application is
creating a centralized database of these historic data to be accessed anywhere over the internet.
11
1.4.1 Target Users
This application is designed to cater to spatial scientists, historians, journalists, and urban
planners that are interested in studying the difference in land use over time in Hong Kong. It is
intended to be a way to quickly visualize what the possible differences are and then take that data
for further use if desired. Even though the primary target users are researchers, it is designed
simply enough to be used by anyone who has a general interest in the subject.
An example of a target user could be an urban planner who is considering a design of a
new section of reclaimed land. The planner could use the platform to see if there is a pattern of
how land use has been transforming near the area of the newly reclaimed land. For example, if
there is a pattern that a neighboring area has been evolving from business and commercially
dominated to more residential, then the planner may have a better idea of how to proportion the
new reclaimed land to suit the growing demand of a specific land use.
Another example of a target user could be the Planning Department and the Government
Records office. These departments want to disseminate the records they have available to the
public but must follow the formalities of their departments. This application could act as a public
preview to the source maps and provide the general public easy access to data that is on their
shelves.
1.4.2 Study Area
The initial version of this project focuses on a few districts on the north and northwestern
coastline region of Hong Kong Island. The districts include Kennedy Town, Sai Yin Pun,
Sheung Wan, and Central. The early colonial city of Victoria covered this extent and therefore is
the area of the island that has the longest land use history. The official units of study area are the
12
Tertiary Planning Units (TPU) 114, 122, and 123. Figure 3 depicts the boundary of the three
TPUs as they are currently.
Figure 3. Boundary of TPUs 114, 122, and 123 demarks study area
1.4.3 Data Overview
The key data for this application are historic maps of TPUs. The Tertiary Planning Unit is
a geographical reference system demarcated by the Planning Department of Hong Kong for the
entire territory. TPUs provide a common geographical system for the compilation of statistical
data. For example, the Census & Statistics Department of Hong Kong would prepare and manage
data of the population census based on the TPU. Also, data requirements of the different users
13
can be met by converting data recorded on the basis of TPU in the users’ geographical
information system for their planning and operational purposes (Planning Department 2018).
TPUs are delineated upon the consideration of various factors including the nature of the
geographical features of the area (such as roads, railway lines, coastlines, contours, and
waterways), lot boundaries, and zoning boundaries in the Outline Zoning Plans. TPU boundaries
would be revised from time to time to reflect the latest urban developments including urban
renewal, urban sprawl, and reclamation.
1.4.4 Data Accessibility
The current accessibility of the historic TPU maps is limited to in-person search at the
Government Records Office in Kowloon. The public has access to view these maps on site and
can purchase copies or digital scans on a per piece basis. Unfortunately, there is no way to
preview the maps remotely or online. One of the project’s aims is to increase the accessibility of
this spatial data in an open online data source so that it can be viewed and used from anywhere at
any time by anyone.
1.4.5 Methodology Overview
The main component of this project involves digitizing historic land use maps to create a
new database. Land use maps are first scanned into a JPEG or PNG format and then brought into
ArcGIS Pro and georeferenced to fit the imported map onto the base map in the correct
coordinate system. Once correctly georeferenced, a new feature class of polygons is made by
hand tracing over the imported historical map. The layers are then published in ArcGIS Online as
web maps and are used as the main component of the web app. The web app was built and
configured with widgets using ArcGIS Web App Builder.
14
1.5 Thesis Structure
The following sections describe the related works, methodology, results, and discussion
surrounding the project. In Chapter 2, a literature review is conducted on the topics of Historical
GIS (HGIS) and Open Data, and it concludes with other examples of similar projects. Chapter 3
covers the methodologies used to complete the project. It goes over the requirements and the
goals of the project, the process of acquiring and processing the data, and how the project was
published publicly. Chapter 4 reports the results of the methodologies used and examines then in
more detail. The thesis concludes with Chapter 5, which discusses the project’s strengths,
insights gained from using the application, as well as challenges as ideas for continued
development.
15
Chapter 2 Related Work
This chapter provides a review of studies related to HGIS, Open Data, and Open GIS. The first
section discusses the importance of HGIS in scholarship and how the combination of geographic
information science (GISci) and history can provide insights for both disciplines. Section 2.1.2
provides examples of HGIS research and its influence. The next sections discuss open data and
open-source GIS. A core concept of this project is about providing an upgraded open dataset that
can be used in iterative and collaborative work. The chapter concludes with discussing the
opportunities and challenges of open data and open-source GIS and provides examples.
2.1 Historical GIS (HGIS)
Historical GIS is the combination of the disciplines of history and GISci, wherein GIS
tools and methodologies are used to explore elements of history. Section 2.1.1 discusses the
importance of HGIS in scholarship and how historians have used GIS. Section 2.1.2 covers
examples of HGIS research and reviews other HGIS projects.
2.1.1 Importance of HGIS in Scholarship
The use of GIS technologies in historical research started in the late 1990s and eventually
developed into what is now known as Historical GIS, or HGIS. As Knowles (2008, 3) states,
“[a]lmost every historical document contains some kind of geographical information.” It
therefore makes sense that more and more historians are using GIS technologies and techniques
to uncover spatial relationships, reconstruct places and natural environments of the past, and
extract more value derived from historical sources (Cillis et al. 2021). In the decades since its
advent, HGIS has experienced a rise in interest for historical research.
16
Lawson (2022) identifies four major characteristics emerging in HGIS. Firstly, all
historical research often has significant geographical elements. Second, a lot of the evidence in
research is actually geographical information. Thirdly, Structuring and organizing the evidence
into databases that record location and time is the basis for analyzing it further. And fourth, the
results are often shown in the form of maps. Mapping is an integral part of GIS, but it is
essentially a database technology, and it is the treatment of historical data as information to be
inputted into a database that makes HGIS (Gregory 2007).
HGIS, and indeed all GIS, are able to provide information about what is featured in the
database, and more importantly, where it is located. HGIS allows for the overlaying of different
kinds of data that can help orient the historian to identify patterns. Bodenhamer (2008) says that
GIS is intrinsically integrative where multiple information sources are layered on a map with the
same spatial units.
There is a slight difference between HGIS and narrative history insofar that traditional
historians are essentially interpretive storytellers that try to provide an explanation of events of
the past. But HGIS uses maps and spatial data to illustrate the patterns but not provide an
explanation (Schlichting 2008).
An intention of HGIS is to provide a way to create new insights of geographies of the
past. These new insights can advance historical scholarship by challenging existing historical
narratives and enable researchers to attempt to answer unresolved questions or ask new ones. A
prime example of this is when in 2006, Knowles explored the Battle of Gettysburg in a new way
by creating a detailed Digital Terrain Model (DTM) of the area and performed a viewshed
analysis to discover what an observer would have been able to see from any location on the field.
This new insight allows historians to identify what officers could have seen throughout the battle
17
and how that could have had an influence on key decisions (Gregory 2007). This shows that the
application of GISci by formulating a spatial question and using of GIS technologies to address
it, can lead to an advance in historical understandings.
There are some clear advantages of HGIS in scholarship. One advantage described by
Gregory (2007) is that because spatial data tells us where the data are located, data that could be
seemingly incompatible can be integrated together based on where on the Earth’s surface the
data are located. Similarly, MacDonald and Black (1998) argue that the use of GIS in history is
useful in providing context by bringing a wide variety of data types together, such as textual,
numerical, and visual, all in the correct location in space and time. These advantages of GIS in
historical research are strengthened by the fact that historical data can be visualized with maps,
animations, and virtual landscapes.
Another major advantage of HGIS is the dissemination of information. Conventionally,
historical research is confined to academic journals and textbooks, but HGIS allows for easy
dissemination of complex historical data via the internet. The combination of assembling a wide
variety of historical information, linking it to a common geography and publishing it on the
internet allows the research to reach a wider audience beyond academics. The simplest way to
disseminate an HGIS project is to publish the data on a web map that allows user to zoom, pan,
turn layers on and off and make queries, such is the goal of this project.
HGIS could be considered an ideal tool for maintaining a collective memory of
environments of the past. Because of the integration with computational systems, HGIS
programs can highlight inconsistency and incompleteness of data, thereby encouraging
researchers to keep a high data-quality standard (Siebert 2000). Using GIS for historical research
18
also makes it easier for the researcher to record metadata, further centralizing related information
of the integrative data.
Naturally, there are some disadvantages when using GIS for historical research. A
common issue in any GIS research is that data acquisition and input is very costly and time
consuming, this is only more exaggerated when dealing with huge amounts of data that span a
long period of history (Siebert 2000).
An example of the challenges and limitations of HGIS is described by Berman (2014),
which addresses the challenges of measuring the space of administrative geographies in
historical China when using HGIS. Berman argues that aligning conceptions of bearings and
distance between the past and the present is essential when reconstructing historical geographies,
but often the accuracy of this information cannot be verified, which then can lead to false
conclusions when running calculations in a GIS. To address this, Berman proposes using
network models to represent established relationships between known points instead of using
bounded jurisdiction models, which are more conventionally used in GIS. Berman says that
when dealing with ancient sources, limiting the types of administrative units is preferred. Instead
of reconstructing estimated boundaries implementing a network model to show the
administrative hierarchy of a specific time using known point locations and administrative
relationships is a better use of time (Berman 2014).
2.1.2 Examples of HGIS Research
Visualizing patterns of historical data with HGIS can have an impact on historical
debates. Such is the case with Geoff Cunfer’s HIGS “On the Great Plains: Agriculture and
Environment, which contributes to the debates about what caused the Dust Bowl of the 1930s
(Cunfer 2005). In this project, Cunfer explores the change of 400 counties in the Great Plains
19
region from the 1870s to the late 20th century using agricultural census data. This allowed
Cunfer to examine crop diversity, irrigation patterns, cultivation and grazing areas and
agricultural mechanization. This was visually presented in choropleth maps.
There are a few HGIS projects that are good examples of information dissemination.
Schaefer’s online GIS of nineteenth century England (Schaefer 2004) and the North American
Religion Atlas by Lancaster (2002) are prime examples. The International Dunhuang Project
(2006) from the British Library uses point data superimposed over raster scans of maps. The
project has created a database of over 100,000 documents, manuscripts, and other relevant
materials from between the 5th and 11th centuries that were excavated on the Silk Road near
Dunhuang, China. The database includes coordinates of the location of each artifact. This online
GIS allows users to query the database through a map-based interface (Klemme 2014). The
Sydney TimeMap Project shows the development of Sydney from early European colonization to
the present day (Wilson 2001).
HGIS can also be used as a tool for deeper research, such as monitoring, preserving, and
planning forest landscapes. Clillis, Statuto, and Picuno (2021) present a method to monitor the
dynamics and management strategies in a case study in the Mediterranean region. In the study,
they explain that the territory is transforming at a rapid pace due to socioeconomic and climate
reasons, which is pushing the ecosystem to be unbalanced. Their goal was to evaluate the
temporal dynamics surrounding this. They used a methodology based on HGIS that allows
different types of geodata to be integrated together. Using historical cartography and remotely
sensed data, they assessed where and how much the forest landscape had changed. Their results
show that the forest area is returning to how it was recorded in the first half of the 1800s with a
20
progressive increase since 1950 due to artificial reforestation and the re-naturalization of forests
due to a decrease in agricultural activities.
HGIS has even been used for health-related studies. Oxford et al. (2001) look at the
correlation between poverty and mortality rates in London from the late 19
th
century to the
1990s. Charles Booth originally conducted a massive survey of the social and economic
conditions of residents of London in the late 19
th
century and constructed large, detailed maps of
the social class of London on a street-by-street level. The result was a clear geography of poverty
and affluence of the period. Using GIS, Oxford et al. combined Booth’s data with more
contemporary ward boundaries and other census data from the 1990s and created an index and
compared the geographies of poverty and mortality rates (Oxford et al. 2001). The study
discovered that the geography of poverty and affluence from the late 19th century was similar to
that of the late 20th century, showing that the spatial patterns of poverty in London have not
changed in a hundred years. Another interesting discovery of the study was finding that causes of
death for people over 65 years old was more closely related to the geography of poverty from the
19th century than in the contemporary period (Oxford et al. 2001).
HGIS has been used to analyze rural landscapes. Statuto et al. identifies land use changes
in a rural area in the Basilicata Region of southern Italy by using territorial analysis with GIS.
Historical maps from 1829 to 2013 were imported and DTMs were used to make 3D
reconstructions of the landscape from different time periods (Statuto et al. 2017). This is another
example where it is imperative to digitize historical cartography to do proper analysis of rural
land changes and understand how land transformations are connected to both human and natural
events. In the 1820s, the study area was mainly dominated by woods and grazing was the main
agricultural activity. By 1876, the landscape was modified to increase agricultural land and the
21
wooded areas were cleared for economic opportunities. The area that was once a semi-natural
zone transformed into a full agricultural area from 1876 to 1955 due to the increase in the
population’s economic needs. The human activities that caused this transformation are now
better understood with this HGIS analysis (Statuto et al. 2017).
Historic population change can be studied in HGIS. McLeman et al. looks at the
historical impact of droughts had on population change on the Canadian Prairies during the
1930s and how there is a complex intersection of climate, agriculture, and population change
(McLeman et al. 2011). GIS was used to map “hotspots” where there was severe heat and lack of
precipitation let to a decline in population in rural areas. The model used digitized census data
combined with historical climate data at 10 kilometer grid cell scale and showed that drought
conditions, decreased prices of commodities and economic recession led to the displacement of
hundreds of thousands of people. The authors suggest that a focus on historical data can provide
a better understanding of how populations have adapted to extreme climate conditions and how
that can provide feedback for future strategies (McLeman et al. 2011).
2.1.3 Examples of HK HGIS
Land reclamation is major part of Hong Kong’s urban development history.
Approximately 6% of Hong Kong’s total area is reclaimed land and the government says that
this accommodates 27% of the city’s population and as much as 70% of all commercial activities
(Vetter 2018). Land reclamation in Hong Kong began as far back as 1841 and over the course of
its colonial history and beyond, more projects have been added to create the final coastline it has
today. In 2014, Ngo et al. developed a web map to visualize the expansion of Hong Kong’s
coastline by land reclamation. Prior to the Hong Kong Coastline Web Map, maps of Hong
Kong’s reclaimed land boundaries were scarce and became quickly obsolete as new projects
22
continued. A map reading group of volunteers spent a year researching historic maps, photos,
and articles to create illustrations of the land reclamation projects throughout Hong Kong’s
history. These illustrations were digitized and converted to spatial data using Google Maps
Engine and then transferred to CartoDB for customization (Ngo et al. 2014). The interactive web
map allows users to visualize the location and progression of land reclamation through time.
Pop-up windows can be opened to provide more information about the data. A slider tool allows
users to animate the land reclamation development as well as narrow the search to specific time
frames. The map also includes geo-tagged historic photos.
Another Hong Kong centered web GIS project is HKMaps.hk. HKMaps.hk was
developed by Tymon Mellor, an engineer that has worked on many infrastructure projects in
Hong Kong. HKMaps.hk is a simple web map app that has a map viewer with and
OpenStreetMap basemap. The viewer has zooming and panning functionality and a few basic
widgets. The main widget opens a list of historical maps that when selected, display over the
basemap. A transparency slider widget allows the user to increase the transparency of the
overlaid historical map to see how it compares with the current map. Other data can be selected
and layered onto the web app such as coastline line data, aerial imagery, and point data of
military and mining sites. HKMaps.hk also includes a tab with a list and preview of all the maps
available in the web app, as well as more information and links to the source.
2.2 Open Data and Open Source GIS
A foundation of this project is to provide an open database of the land use maps sourced
from the Hong Kong Government records office. This section discusses open data and open-
source GIS. Section 2.2.1 covers the definitions of open data while section 2.2.2 covers the
origins of the open data movement. Section 2.2.3 describe how open data related to GIS and
23
geospatial data. Section 2.2.4 covers the opportunities and challenges surrounding open data
while section 2.2.5 provides examples of open data projects.
2.2.1 What is Open Data?
The definition of open data is data which is not restricted by copyright laws and is user
accessible, meaning it is available using platforms or tools that are free and easily accessed by
the public. The Open Knowledge Foundation (2023) defines open data as “data that can be freely
used, shared, and built-on by anyone, anywhere, for any purpose.”
Licenses are used in the field of open data for organizations to set parameters on how the
data may be used or accessed. This usually includes legal information regarding the data and
how it can be used. Common licenses such as Creative Commons exist while government
agencies and organizations have developed custom licenses that explain the parameters of how
the data may be re-distributed or used. Open data can be organized at the country, state,
organization, or topic level (World Bank 2021).
The ideal for open data is that it is useable, and for this to occur, data should have
complete metadata. Two indicators that open data is accessible and being successfully used are
when open data leads to innovations within a particular field and open data is incorporated into
existing or new projects or products. The hope is that as open data’s value is seen and recognized
and utilized within new business, products, and projects and this reinforces the demand for open
data and therefore promotes more open data to be created (World Bank 2021).
Open data stems from the concept of open science, which refers to science conducted in a
collaborative environment with a focus on research communication. Both open science and open
data consist of principles of transparency, reuse, participation, accountability, and practices of
using open publications, sharing data, and citizen science. (Charalabidis et al. 2018).
24
Wilkinson et al. (2016) created the acronym FAIR, which stands for findable, accessible,
interoperable, reusable. They (2016, 4) argue that “FAIRness is a prerequisite for proper data
management and data stewardship” and “good data management and stewardship is not a goal it
itself, but rather a pre-condition supporting knowledge discovery and innovation”.
2.2.2 Origins of Open Data
Simon Chignard, an expert of data governance, describes the open data movement as a
techno-political idea that has rapidly spread around the globe. He underpins this statement with
the definition of open data being data that must be free technically (accessible using free tools),
legally (little to no copyright barriers), and economically (free without paywalls).
Chignard mentions that this idea is not something new and that the scholar Robert Merton
was already advocating that all science research should be publicly accessible in 1942 (Chignard
2013). The Nobel Prize of Economics in 2009 was awarded to Elinor Orstrom, who unveiled
how a “common good” was technically possible. Orstrom showed that not only does open data
allow multi-user access (meaning nobody’s work is impeded by the work of another) but also
that the open data idea can actually lead to innovation or further research being added to the
common good (Ostrom 1990).
Open data was first used in an environmental government publication in 1995 to collect
and utilize data to compile a more complete and global image of the state of the biosphere and
oceans (Chignard 2013).
2.2.3 Open GIS
Open data and open-source software have been making a significant impact in GISci.
Over the past few decades, GIS has witnessed significant growth encompassing various
proprietary and open-source software. Free and Open-Source Software for Geospatial (aka
25
FOSS4G) is a recurring global event held since 2006, encouraging and promoting open-source
software development (Maurya et al. 2015). GISci has seen an increase in desktop, mobile, and
web applications and the open-source movement has played a critical role in this growth. The
criteria for open-source software is that it must have be free to redistribute, contain the source
code, include derived works, protected the integrity of the author’s source code, does not
discriminate against any persons or groups, does not discriminate against any field of endeavor,
the license must not be specific to a products and the license must not restrict other software
(Maurya et al. 2015).
Open-source web GIS has been used for monitoring urban land use planning.
Development of a Web GIS application framework using open-source software to provide open
data services for policymaking and urban land use planning. The application was developed
using a comprehensive geodatabase model and user interface using open-source tools such as
QGIS, MapSever, and P mapper. The system is designed to create a complete digital map service
with information on urban land use policy, suitable for public consumption and a decision
support system for stakeholders (Sejati et al. 2020). This web GIS framework was applied to
Balkipapan City, in East Borneo, Indonesia. The expectation of the system is to serve as a model
for land-use monitoring based on the principles of information disclosure towards smart city and
smart governance. The framework was developed to mainly monitor urban land use planning in
developing countries and includes a user interface with menus for accessing data such as
infrastructure planning, search, and reporting. Balikpapan’s Web GIS menus uphold the
principles of open data, not only enabling operators but also the public to report on infrastructure
development and land use. The display data is sourced from both the government appointed
operators and the public, this is to stay in line with the principles of smart governance which
26
include data transparency and participatory data. The authors state that the implementation of
such a framework can have a positive impact on public trust in the government. The use of this
web GIS is a step towards bridging the communication gaps between the public and government
(Sejati et al. 2020).
Open-source is also used in database building. Open-source databases and their spatial
extensions provide robust and reliable solutions for managing spatial data (Chen 2008). As the
demand for spatial data handling increases, open-source databases are likely to see more
development and adoption. Currently, MySQL is the most extensive open-source database
software and is available on operating systems such as Linux, Mac, and Windows. PostgreSQL
is highly scalable, is SQL compliant and has a strong reputation for reliability, data integrity, and
correctness. PostgreSQL has a spatial extension called PostGIS, which conforms with OGC and
ISO SQL/MM standards. This makes it possible to use PostgreSQL for modern spatial
application development. (Chen 2008).
Open-source geospatial software is regularly used by governments, business,
professionals, and academics and the communities of open-source geospatial software and open
geospatial data highly intersect. The Open Geospatial Consortium is closely linked with open
geospatial standards and ensures geospatial interoperability (Minghini et al. 2021). The
relationship between open-source geospatial software and open geospatial data is likely to get
even closer.
2.2.4 Open Geospatial Data
There are three types of open geospatial data: collaboratively contributed data,
authoritative data, and open scientific data. Collaboratively contributed data is when volunteers
collect and maintain geospatial data, usually by crowdsourcing. Authoritative data is collected
27
and shared by governments and other public administrations to promote the freedom of access to
information. Scientific data is shared to help with the verification of research findings and to
integrate results (Coetzee 2020).
Collaboratively contributed open geospatial data are distinguished between user
generated content, crowdsourcing, and citizen and community science. Open Street Map (OSM)
is the most widely known example of an open geospatial dataset that is maintained and expanded
by its community of contributors. Raifer et al. presents the OpenStreetMap History Database
(OSHDB), which is a framework for spatio-temporal analysis of OSM history data. This allows
for spatio-temporal analyses of OSM data on a global scale, a user-friendly way to create large
scale visualizations (Raifer et al. 2019). But there is still an issue with the data quality of OSM, it
may vary in accuracy, completeness and precision since the enthusiasts that contribute might
have different GIS knowledge and skills (Coetzee 2020).
Authoritative data is when governments collect and maintain geospatial vector data such
as administrative boundaries, building footprints, street centerlines, etc for governance and
management purposes. The idea is that open and shared authoritative data could lead to more
efficient and effective decisions for government. A specific example of this is the European
Amended Public Sector Information Reuse Directive, which aims to make public government
data available for reuse with fewer legal restrictions. The sharing of open public data can
increase entrepreneurship and innovation. Spatial data infrastructure (SDI) facilitates the sharing
of spatial data among the different stakeholders in the community, and an open SDI is not just
about providing open data, but also organizing the infrastructure around it to encourage the
participation of others outside the government (Coetzee 2020).
28
Open scientific data allows for scientific claims to be verified, confirmed, or rejected in
an open way. The need for open scientific data has been recognized for over fifty years by the
International Council for Science. The promotion of FAIR principles aims to improve the
sustainable use of digital resources such as data. Scientific journals such as Nature and Scientific
Data endorse the notion of science as an open enterprise and will only publish articles with
supplementary material that meet FAIR standards (Coetzee 2020).
2.2.5 Opportunities and Challenges
The open GIS concept includes 8 dimensions: data, software, hardware, standards,
research, publication, funding, and education. Open GIS offers 4 opportunities for the GIS
community: technology-driven opportunities, application-led opportunities, crowd powered
opportunities, and education focused opportunities (Sui 2014).
Despite it becoming increasingly important in shaping future research and educational
agendas, there are academic, legal, social/political, and environmental impediments for the
practice of open GIS (Sui 2014).
The future of the open-source communities is uncertain. Open data can be difficult to
ensure quality, and openness makes the data and software more vulnerable to abuse and
manipulation. Both open-source projects and open datasets are examples of shared resources that
are maintained by the community, so there are questions around how contributing volunteers can
be retained (Coetzee et al. 2020).
Although greater transparency and accountability, improved data sharing, increased
citizen access to government decision-making is a positive effect of open data, there are also
some downsides. Open data that fails to address privacy concerns may lead to negative and
harmful outcomes, such as the emotional, financial, and physical targeting of individuals
29
(Johnson et al. 2017). There is a danger when governments provide open data to support
economic development because they might be supporting a model that promotes private sector
services instead of using open data for citizen-facing services. Private sector companies may use
open data to gain commercial advantages, such as bidding to supply public sector services.
Private sector companies may also repackage, process, and analyze government data for the
government or private sector. Taxes paid by citizens to fund the provision of open government
data are affected by the indirect costs of not having reliable access to the data. At the same time,
private sector companies profit by offering a private service back to the public for a fee.
Therefore, the government needs to be cautious of using open data to support economic
development that promotes private sector services instead of citizen-facing services (Johnson et
al. 2017).
There is a cost to geospatial open data. The indirect costs are citizen participation
challenges, uneven provision across geography and user types, subsidy of private sector
activities, and corporate influence on government. These indirect costs may stunt the
effectiveness of open data in increasing transparency and citizen engagement in government.
Johnson et al. suggests that governments need to manage the indirect costs of open data while
maximizing its positive effects. It must balance competing values such as privacy and
transparency and develop outreach programs to educate the public on how to access, use, and
interpret open data.
Open geospatial software and open data will influence each other in the future. The
increase in the use of spatially enabled services and shifting towards software that can process
spatially-enabled data (Coetzee et al. 2020). Because of the increasing volumes of free and open
geospatial data, there is a need for computing platforms that can extract useful information from
30
the data in an efficient way. Cloud computing platforms are changing the way geospatial data are
processed, analyzed, and visualized (Coetzee et al. 2020).
2.2.6 Urban and Land Use Open Data/GIS
GRASS GIS is a muti-purpose open-source GIS. Geographical Resources Analysis
Support System (GRASS) is a key open-source GIS that has been under development for over 28
years. It is a mutli-purpose GIS that can be used for geospatial data production, analysis, and
mapping. The development of GRASS is community based, the developers are distributed
globally, communicating through online source code repositories, mailing lists, and a wiki
(Neteler et al. 2012). The latest release of GRASS has more than 400 modules and natively runs
on all popular operating systems, giving basic and advanced functionality to users of all kinds.
GRASS offers many spatial modeling algorithms, 3D visualization, image processing routines,
and integrates well with other open source and proprietary software packages that have GIS
applications. GRASS has been applied in the field of public health, epidemiology, and infectious
diseases (Neteler et al. 2012).
There is increasing importance of metadata in the GIS community, especially in the
academic context. It is becoming more complex to create and maintain metadata. There is
preliminary work on an approach to overcome these issues using the Free and Open Source
(FOSS) GIS products quantum GIS 1.8.0 and PostreSQL 9.2 with Post GIS 2.0. The authors
describe how metadata creation can be automated. It couples metadata and data, and ensures that
the metadata is updated automatically when the data changes. The motivation is to automate
metadata creation that can save time and effort, improve metadata quality and consistency, and
contribute to interoperability among multiple GIS (Ellul et al. 2013).
31
There is a study that linked open geospatial data for predicting urban semantics and land
use classification in Europe. OSM/Linked geodata was used to extract features related to land
use, land cover, and infrastructure (Re Calegari et al. 2015). The study found that linked open
geospatial data can be used to support sustainable urbanization and planning activities is less
costly than manual work.
There is potential of using linked open geospatial data in extracting actionable knowledge
about the urban environment to support sustainable urbanization processes and offer innovative
solutions to smart city problems (Steiniger 2009). The article presents set of experiments that use
geo-information about points of interest as input in a classification model of land use over wide
urban areas in Europe. The experiments were replicated in different European cities to ensure
replicability. This emphasizes the importance of monitoring changes in land use to support a
sustainable urbanization process. The experiment aimed to develop a classification model for
land use in four European cities, Milan, Barcelona, Munich, and Brussels. The article discusses
the use of linked open geospatial data in predicting urban land use, with the aim of supporting or
updating expensive spatial data sources. The authors suggest that combining complementary
geospatial information from OpenStreetMap with other sources like phone activity data, could
lead to better solutions (Steiniger 2009).
32
Chapter 3 Methodology
This chapter reviews the requirements of the project and the methodologies for completing it.
This chapter begins with a review of the project’s requirements and objectives, which are the
guideposts of the application’s development. The following sections cover the steps necessary to
develop the application. Section 3.2 covers the software choices needed to execute the
methodologies and deploy the final version. Next, section 3.3 discusses the data used in creating
the new database. Finally, section 3.4 covers the process of building the historical database and
application.
3.1 Requirements and Objectives
The objective of this application was to create a platform that allowed users to visualize
historic land use data in the central district of Hong Kong and to provide access to this data. This
section outlines the requirements that were necessary to fulfil this objective. A goal of the
application is to display the land use survey data of the tertiary planning units (TPUs) and allow
users to see the different land use types across years. By using a digital map, users can see
multiple survey map data at the same time and pull this data for their own use. To accomplish
these objectives, the necessary requirements are explained below.
3.1.1 Accessibility
The accessibility requirements for this application prioritize easy access. The application
must be available to users via a desktop or mobile device. There should be no additional program
downloads or account setup for most of the functionality, therefore users will only need a default
web browser to access the application. Lastly, the application will require no training to use,
making the application accessible to anyone regardless of GIS background.
33
3.1.2 Functionality and Intended Users
This application is designed to provide specific functionality to meet the needs of various
intended users. Firstly, the app must have the capability to visualize the georeferenced land use
types of the TPUs. Users should be able to explore the data by panning and zooming using a
mouse or trackpad. The application should also allow users to toggle layers for different TPUs
and years, enabling them to selectively view the desired information. Additionally, users should
have the ability to import and visualize other data alongside the land use data, enhancing the
analytical capabilities of the app. There should be search and query functionality to facilitate the
identification of specific land use types. Lastly, the application should support data export
functionality, allowing users to utilize the data in other software programs for further analysis.
The intended user base for this application is diverse, catering to a range of disciplines
and interests. The primary intended user is an academic researcher who is interested in exploring
the spatial history of Hong Kong. The app and accompanying dataset can serve as a valuable
resource for analyzing the transformation of land use in Hong Kong from its colonial era to its
current status as a part of China. GIS professionals and academics specializing in the field can
also find this application and associated thesis helpful as a reference for developing similar
historical GIS projects.
Furthermore, key users of the application could include the Hong Kong Public Records
Office and the Hong Kong Planning Department. The Public Records Office houses the original
paper maps of the TPU land use surveys that have been digitized for this project. Researchers
interested in accessing these maps currently need to physically visit the office, making the
research process time-consuming. The website of the Public Records Office catalogues the maps,
but without preview images, making it difficult to ascertain the contents of each map without a
34
full download. The availability of a readily accessible website application like this one would
significantly reduce the time and effort required to study these land use maps.
Finally, the application also aims to serve the general public and non-professionals who
have an interest in Hong Kong's history. History enthusiasts and curious individuals would find
value in an application that provides access to historical data, offering insights and perspectives
that might not be readily available otherwise.
3.1.3 Software Requirements
Since this application is an early version that will continue to be developed over time, the
software requirements must reflect this. There are four main things when considering which
software to use to develop the application. First, it must be capable of producing web maps.
Second, it must allow for free sharing and access for the user. Next, it must include tools to
interact with the presented data. And lastly, it must allow for continued development for future
iterations.
3.2 Software Choice
For better integration of the software requirements, a software platform that could handle
each part of the requirements was preferred. Esri is an American GIS software company that is
best known for its ArcGIS products. ArcGIS products are considered the gold standard of GIS
software and most notably include ArcGIS Pro and ArcGIS Online. ArcGIS products are capable
of spatial analysis, mapping, data collection and management, imagery and remote sensing, and
3D GIS (Esri). Esri’s ArcGIS products are used by governments and universities worldwide. The
Esri product suite fulfills the map digitizing requirements with ArcGIS Pro, the publishing
requirements with ArcGIS Online, and the customization requirements with ArcGIS Web
AppBuilder.
35
ArcGIS Pro is the software heavily used at the Spatial Sciences at USC. It is Esri’s
professional desktop GIS application and has full-functionality for building maps and visualizing
data. For the first part of this project, ArcGIS Pro was the appropriate software to bring in
scanned images of the historical TPU maps and convert them to digital polygons.
ArcGIS Online is a cloud-based mapping and analysis program. It is not as robust as
ArcGIS Pro but it is preferred for its sharing capabilities. With ArcGIS Online, content can be
shared inside and out of an organization and is a good choice for collaborative projects. For this
project, sharing the polygon layers built in ArcGIS Pro to ArcGIS Online was a necessary step
towards building the web map application.
ArcGIS Web AppBuilder is application available to run via ArcGIS Online. It is an
application that allows users to build unique and dynamic experiences without having to write
any code. This is the preferred publishing software for the project as it is already seamlessly
integrated with the other Esri software.
3.3 Data
The data used in this project are historical land use maps of TPUs. Tertiary Planning
Units are a geographical reference system that subdivide Hong Kong’s territories into units
smaller than districts. The six maps used to create spatial data are from TPU 114, 122, and 123.
These TPUs were chosen as they are clustered together in neighboring districts. There are two
maps from different years for each TPU as is shown in Table 1. Maps created before 1980 used
the Hong Kong 1963 Grid coordinate system, while maps created after used the Hong Kong
1980 Grid. A description of each historical TPU map is given in sections following.
36
Table 1. Land use survey map data
Dataset District Source Date Scale Coordinate System
TPU114_1972 Sheung Wan HK GRS Nov 1972 1:2400 HK 1963 Grid
TPU114_1982 Sheung Wan HK GRS April 1982 N/A HK 1980 Grid
TPU122_1967 Central HK GRS Aug 1967 1:2400 HK 1963 Grid
TPU122_1985 Central HK GRS Aug 1985 1:2500 HK 1980 Grid
TPU123_1967 Central Area HK GRS Aug 1967 1:2400 HK 1963 Grid
TPU123_1985 Central Area HK GRS May 1985 1:2500 HK 1980 Grid
3.3.1 Map of TPU 114 (1972)
TPU 114 is located in the middle of the Sheung Wan neighborhood, on the northwestern
part of Hong Kong Island. Sheung Wan is the neighborhood directly west of the main Central
District. Part of TPU 114 sits along the coastline with the harbor. The map of TPU 114 from
November 1972 shown in Figure 4 is at a scale of 1:2400. The original map has faded to a
yellowish color and shows wear and tear around the edges. This map depicts TPU 114 at a time
when it met with the coastline, showing protruding ferry piers.
37
Figure 4. Original map of TPU 114, year 1972
3.3.2 Map of TPU 114 (1982)
The map of TPU 114 from April 1982 as shown in Figure 5, does not have a specified
scale. The extent of the map is smaller than the 1972 version. The map is less faded and shows
streets and building plots more clearly. The survey data of this map does not include the
coastline area and ferry ports as it did in the 1972 version.
38
Figure 5. Original map of TPU 114, year 1982
3.3.3 Map of TPU 122 (1967)
TPU 122 borders TPU 114 to the east and is smaller. It is further inland away from the
harbor and closer to Mount Austin. This TPU mainly revolves around the parcel that was the
Central Police Station. The original map for TPU 122 from August 1967 as depicted in Figure 6
is at a scale of 1:2400 and shows signs of fading within the legend.
39
Figure 6. Original map of TPU 122, year 1967
3.3.4 Map of TPU 122 (1985)
The map for TPU 122 from August 1985 is at a scale of 1:2500. Figure 7 shows that the
extent of the map is different from the 1967 version, it includes more depiction of Mount Austin
in the southwest and less of the northern coastline.
40
Figure 7. Original map of TPU 122, year 1985
3.3.5 Map of TPU 123 (1967)
TPU 123 is in the middle of the Central district and shares a border with TPU 122 to the
west. TPU 123 includes the Government House, which was the office and residence of the
Governor when it was a colony and for the Chief Executive after. This TPU once sat along the
harbor before more land was reclaimed. The map for TPU 123 from August 1967 is also in a
41
scale of 1:2400. It highlights the land use from Government House down towards the coastline,
as is shown in Figure 8. As one of the older maps of the series used in this project, the paper is
yellow from age.
Figure 8. Original map of TPU 123, year 1967
3.3.6 Map of TPU 123 (1985)
The map for TPU 123 from May 1985 is at a scale of 1:2500. Scales of all maps used in
the project are referenced in Error! Reference source not found.. As seen in Figure 9, the 1985
42
map of TPU 123 shows less extent and includes land use data of the ferry pier, which was visible
in the 1967 map but not categorized.
Figure 9. Original map of TPU 123, year 1985
43
3.4 Building the Historical Database
The construction of the historical database was the most time intensive component of this
project. Ultimately, the utility of the application and the data it contains relies on the accuracy of
transforming the historical maps into digital data. Section 3.4.1 describes the georeferencing
process that aligned the historic maps to a coordinate system, and section 3.4.2 covers the steps
taken to create a digital layer of each map’s data.
3.4.1 Georeferencing
In many GIS and HGIS projects, the building of the database is considered to be the most
time-consuming stage, and such was the case in this project. To capture vector data from the
historic maps, the process of digitizing was initiated. Captured data from a map or another source
like aerial imagery is considered as secondary data capture because the source is at least one
stage removed from the real world (Gregory 2007). With paper maps, they first needed to be
converted to a digital image by using a high-resolution scanner. Since the TPU maps had already
been digitized by the Government Records Office, they were ready to be brought into ArcGIS
Pro for further processing. The maps were imported as PNG files to keep high resolution.
Georeferencing is a fundamental GIS technique which converts the raw coordinates of
the image into a real-world coordinate system. When the layer has real-world coordinates, it
means it can be used to calculate distances, areas, and angles in real-world units of measurement.
Also, the layer can then be integrated with other layers that have real-world coordinate systems
(Gregory 2007). Georeferencing is done by adding control points to the layer. In technical terms,
it refers to the alignment of spatial data on a geographic grid (Skopyk 2021). The Hong Kong
1980 Grid was chosen as the as the geographic grid for this project. The Hong Kong 1980 Grid is
44
a projected coordinate system widely used in Hong Kong for describing position of land
boundaries, map features, buildings, and town planning zones (Kwok 2012).
When bringing a map image into ArcGIS Pro, it will not automatically display in the area
of interest on the map; it has no spatial reference. To begin the process of digitization, the correct
coordinate system of the basemap needed to be specified. To match the coordinate system of the
historical paper maps, the Hong Kong 1980 Grid coordinate system was selected. Each map then
needed to be brought into the approximate extent before it could be further georeferenced. To do
this, the background reference map in ArcGIS Pro was roughly zoomed in on the location of
central Hong Kong. The map image layer was set to display.
Once the map image was displayed in the correct approximate extent, control points
could be added. Typically, four or more control points are used and positioned in the corners of
the map sheet. Control points were added by activating the Georeferencing ribbon in ArcGIS Pro
and selecting Add Control Points. First, a point on the digitized historical map was selected, then
another point was selected on the map in the GIS to rectify where the original point should have
been. It is important to understand the map and the study area when choosing appropriate
control points. In the case of the TPU maps, they had all been produced between 1967 and 1985
and many of the building lots had changed. To select control points that were consistent between
the historical map and the modern background map, major street intersections were chosen as
well as corners of the Government House parcel. After the input of each control point, ArcGIS
Pro automatically bends, stretches, and rotates the pixels in the map image to fit the geographic
grid.
When the georeferencing transformation happens, a mathematical equation is assigning
XY coordinates to every pixel of the map image. In ArcGIS Pro, there are several transformation
45
options, and each has a unique advantage. The zero polynomial simply shifts data, first order
polynomial optimizes for global accuracy, and the 2nd and 3rd order polynomial provide local
but not always general accuracy (Skopyk 2021). For this project, first order polynomial was
chosen. Once the georeferencing was completed with control points, it was saved as a new raster
layer and each layer was set to the Hong Kong 1980 Grid coordinate system.
3.4.2 Digitizing Historical Maps/Creation of Spatial Data
With the maps accurately georeferenced, the process of digitizing each map could begin.
Digitizing the map effectively means to create spatial data of the content within the map. A new
file geodatabase was created in ArcGIS Pro and from there a new feature class was created for
each georectified TPU map. When creating each new feature class, the two new text fields were
added to the attribute table for Type and Subtype; the original maps had general land use types
and then more specific types under the general category. The feature class was set to the same
Hong Kong 1980 Grid. The new feature class created a new layer in the project, which was then
selected and in the Edit tab, the Create function was chosen to open the Create Features pane.
From the pane, the option of creating polygon features was chosen. Using this tool, each polygon
on the original map was traced by hand. After each polygon was drawn, attribute values were
typed in for the zoning type and subtype.
3.5 Web App Development
After the spatial data had been created, the next phase of the project was the development
of the web app. Using chosen software, workflows were followed to make the spatial data
publishable and layout designs were selected. Section 3.5.1 discusses the publishing of the web
layers. Section 3.5.2 covers how the web map was designed. Section 3.5.3 describes the building
of the web app and lastly, section 3.5.4 covers how the web app was launched and published.
46
3.5.1 Publishing Web Layers
Once the maps had been digitized in ArcGIS Pro, the next step was to publish the layers
so they can be used in web map. To do this, each layer first needed to be adjusted to be in the
WGS 1984 Web Mercator (auxiliary sphere). This was done by changing the coordinate system
in the layers’ properties. The layers were then selected and chosen to Share As Web Layer. After
configuration and an analysis of potential publishing errors, each layer was then published to
ArcGIS Online and listed under Content.
When sharing as a web layer, a layer type must be selected from either feature type, tile
type, or vector tile type. Web feature layers support vector feature querying, visualization and
editing. This is the option most suitable when visualizing data on top of a basemap. A web tyle
layer uses predawn map images (called tiles) for fast map visualization. This layer type is best
for basemaps. Vector tile layers are vector tiles and style resources that adapted for customized
display. These are best for basemaps or operational layers. Since the HK HULU app prioritizes
showing the data on top of a basemap, feature layer types were chosen when publishing the web
layers.
3.5.2 Designing the Web Map
A new web map was created in ArcGIS Online to consolidate all the digitized polygon
layers from ArcGIS Pro. In order to focus attention on the digitized polygon layers, an
appropriate basemap needed to be chosen that would adhere to cartographic best practices. The
Light Grey Canvas basemap was selected as it offered enough detail to view the polygon layers
in context but simple enough that it did not distract the viewer from the important information.
To complement the choice of basemap, the symbology for the layers needed to be
colorful but not busy. A problem occurred when originally setting the symbology in ArcGIS Pro.
47
When the layers were symbolized using the unique values choice in the symbology pane, the
colors were unique based on attributes Type and Subtype. However, when this was applied to
other layers, the unique values colors did not match the ones of the previous layer. This was
because not all layers had the same combination of Types and Subtypes included in their
attributes. To solve this, a new field in the attribute table was created titled Type_Subtype so
each combination of Type and Subtype would be its own unique value. The unique value
symbology was set to this new field. There were 25 unique combinations. To complete the
symbology and make sure that the same colors would be used in each layer, 25 unique RGB
values were generated by best matching the color scheme from the source maps and applied to
each Type_Subtype in all layers. The Warehouse and Storage and Misc(Car Park) categories
were symbolized with hatching on the source maps but that functionality was not compatible
with ArcGIS Web App Builder so a solid color that best represented the category was chosen.
The details of RGB values and HEX code for each Type_Subtype are shown in Error!
Reference source not found..
48
Table 2. RGB values for land use categories
Type/Subtype R G B HEX # Color
Business and Office 255 170 0 FFAA00
CIP 245 245 220 F5F5DC
Commercial 241 70 106 F1466A
Educational 240 230 0 F4E600
Educational (Government) 248 220 15 F8DC0F
Educational (Private) 248 237 18 F8ED12
Industrial (General and Heavy) 160 100 180 A064B4
Industrial (Light) 213 165 219 D4A5DB
Institution and Community (Government) 108 177 158 6CB19E
Institution and Community (Private) 138 252 185 8AFCB9
Misc (Car Park) 255 180 100 FFB464
Misc (Other Unused Land) 240 240 230 F0F0E6
Misc (Vacant Building Land) 248 248 245 F8F8F5
Private Open Space (Active) 170 255 150 AAFF96
Private Open Space (Passive) 106 241 119 6AF177
Public Open Space (Active) 160 255 100 A0FF64
Public Open Space (Passive) 129 241 82 81F152
Residential (Apartment Buildings) 183 166 110 B7A66E
Residential (Govt. & Govt. Aided Housing) 70 45 29 462D1D
Residential (Tenement Buildings) 204 191 139 CCBF8B
Residential (Zone 1) 219 205 153 DBCD99
Residential/Commercial 255 120 150 FF7896
Special Categories (Other Uses) 230 187 187 E6BBBB
TBD 128 128 128 808080
Warehouse and Storage 130 70 150 824696
49
3.5.3 Designing the Web App
The Web App is the final published version of the project application. Because it was
built using ArcGIS WebApp Builder, it did not require any code to design the specific elements
of the application. However, design choices were made regarding layout and widgets were
chosen and configured to execute the required functionality of the application.
3.5.3.1 Layout Design Choices
First, the Theme of the app had to be selected. A Theme in WebApp Builder is an out-of-
the-box layout template that gives the app it’s look and feel. Themes are organized by panels,
styles, and preconfigured widgets. It is important to select a Theme with the end-user experience
in mind. There are several Themes to choose from when working with ArcGIS WebApp Builder,
Themes include: Billboard, Box, Dart, Dashboard, Foldable, Jewelry Box, Lanchpad, Plateau,
Pocket, and Tab. All of the Themes are pre-designed for a few general purposes. The Billboard
theme is the most basic which is for apps with very simple tasks. Other themes such as the
Dashboard theme is more focused on showing all widgets open at the launch of the app. The
Pocket theme is designed for apps that are intended to be embedded in other websites.
The Hong Kong Historical Urban Land Use application is a map-centric app. The main
focus on the application is the layers of spatial data that had been digitized from historical maps.
The Dart theme was chosen as it provides a wide view of the map. The application is also meant
to be engaged with, allowing users to view different layers and perform other actions to gain
more insights from the data. Even as a map-centric application, the app also needed to support a
host of widgets that give users easy access to tools to analyze the data further. The Dart theme
has a ribbon that runs along the bottom of the screen where the widgets were brought in neatly so
all functionality was available without disturbing the attention of the map. The Dart theme also
50
allows for widgets to be opened in smaller windows on the screen that can all be opened at the
same time and moved around the page, giving the user flexibility without over complicating
potential workflows.
3.5.3.2 Widget Choices
Widgets are tools that control the functionality of an app in ArcGIS WebApp Builder.
The Hong Kong Historical Urban Land Use application is meant to be interacted with, and the
spatial data available for further analysis and processing. There are many different widgets
available when creating a project and each Theme comes with a few pre-configured widgets. The
Dart theme automatically loads with map zoom buttons, a legend widget, and layers widget. The
legend widget is necessary to identify the land use types in each layer. The layer widget allows
the user to select which layers are visible; this is very important since the first version of this app
includes land use data of the same TPUs from different years, so the layers overlay each other.
The next widget added was the Filter widget. This was chosen so that users can change
how the map is displayed based on the attributes they are interested in. The Filter widget allows
for custom filter choices, an important part of the user experience. Next, the Screening widget
was added. The Screening widget allows the user to identify an area of interest and analyze
specific layers. In the widget, users can identify the area of interest by searching for a place
name, drawing an area, or uploading a shapefile. The Screening widget is a fundamental tool
because it allows for the area of interest to be downloaded in CSV, File Geodatabase, and
shapefile formats.
The Query widget allows for users to retrieve information from the source data. When
configuring the Query widget, the six TPU data sources from the web map were set as the
queries. This allows the user to query each layer and retrieve its information. Queried results
51
have the option to be exported to CSV, JSON, and GeoJSON files. The next widget chosen was
the Select widget. The Select widget is a similar tool to the Screening and Query widgets but the
user can easily use the mouse to highlight or click on the features of the map they are interested
in. Once highlighted then the user has the option to export the selection to CSV, JSON or
GeoJSON form. A key difference between the Select widget and the Screening widget is that the
Select widget does not allow for the download of shapefiles. The Swipe Layer widget was added
as an extra tool. This widget allows users to swipe to see the difference between one layer and
another. Since this application portrays the change of data from the same locations over time, it
is a convenient feature to have a quick visual comparison at the historical land use change.
The next widget added was the Add Data widget. The Hong Kong Historic Urban Land
Use app strives to provide the most amount of functionality for an array of users. It is important
that users can use the app to gain further insights for their own specific purposes. The Add Data
widget allows users to bring other data into the web app. Users can search data from ArcGIS
Online (if they have access), or they can input a URL, or even upload a file in shapefile, CSV,
KML, GPX, and GeoJSON formats.
While not technically a widget, the Attribute Table function was enabled in the app to
allow users to have a more traditional GIS experience. The Attribute Table is a separate tab at the
very bottom of the page, not along the ribbon with all the other widgets.
The About widget opens a pop-up window with information about the application and
provides tips on how to use the widgets. Also, a link is provided to a GitHub page where the full
dataset is available in shapefile format.
Finally, the Share widget was added. This widget is to make it easy for users to share it
with their contacts/audience by providing a link or embedding it on a website.
52
3.5.4 Publishing the Web App
The final process to the project was publishing a finalized web app. Since the web app
pulls the map from the ArcGIS Online, it was first necessary to adjust the settings to make sure
that all the layers could be visible publicly under the Share options. This process had to also be
repeated for web map and web application. Lastly, the web app was published by selecting the
launch button in ArcGIS WebApp Builder.
53
Chapter 4 Results
This chapter displays and discusses the results of the methodologies described in the previous
chapter. The combined result is a working, publicly accessible web application that supports the
functionality of exploring the data layers within it. The application is accessible via the URL:
https://arcg.is/1GHWPW0. The chosen built-in widgets allow for filtering, querying, an
downloading data. There is also functionality to bring in additional outside data into the
application. This chapter opens with a full description of the spatial data created and then
explores the results of the web map, followed by the final web mapping application.
4.1 Root Mean Square of Control Points
Georeferencing can be error prone. If the paper map is folded or creased or not placed
perfectly flat in the digital scanner, or if the control points are inaccurate, the resulting map layer
can be distorted. The cause of the distortion is called the root mean square (RMS) error. An RMS
error is the difference between where the from control point ended up in comparison to actual
specified location (ArcGIS Pro Help n.d.)
The connection between RMS errors and the coordinate system is a crucial one. The
coordinate system provides the reference framework needed to take accurate measurement of the
errors. Because the coordinate system establishes a consistent frame of reference, the positions of
the control points can be properly defined and can help align the measured points with the
expected points.
ArcGIS Pro provides a measure of the RMS errors. Using more control points is likely to
increase the accuracy of the georeferencing and reduce the RMS error. Table 3 shows details
concerning the transformation, number of control points, and RMS error breakdown. Because
54
both the basemap and the historical maps shared the same Hong Kong 1980 Grid coordinate
system, and sufficient control points were used, none of the RMS errors were particularly large.
This indicates a higher level of georeferencing accuracy.
Table 3. RMS Errors Table
RMS Errors
Map Transformation # Control Points Forward Inverse Forward-Inverse
TPU114_1982 1st Order Polynomial 7/7 1.92904 3.790704 0.00000
TPU114_1972 1st Order Polynomial 6/6 1.021159 1.930441 0.00000
TPU122_1967 1st Order Polynomial 6/6 2.02319 3.783216 0.00000
TPU123_1967 1st Order Polynomial 10/10 3.792342 7.172299 0.00000
TPU123_1985 1st Order Polynomial 7/7 1.235924 2.249308 0.00000
4.2 Spatial Data Created
The first goal of this project was the creation of spatial data from images of historic
maps. The creation of this spatial data was the most time-consuming part of the project’s
development. The results offer a better visual experience when observing and a much improved
opportunity for analysis of the content of each particular layer.
The first layer is of TPU123 from year 1967 and is shown in Error! Reference source
not found.. This layer contains 17 of the 25 different land use categories from all 6 historical
maps. The distribution of the categories is relatively even and not one category dominates the
map. The categories are divided into separate quadrants of the TPU. The northwest quadrant
mainly consists of categories Business and Office and Residential (Tenement Buildings). The
center of the TPU is where the Commercial category is located including the
Commercial/Residential category which signifies hotels. The northeastern quadrant mainly
55
consists of Public Open Space (Passive) and Institution and Community with both Private and
Government subtypes. There is one large Private Open Space (Active) category which was the
location of the Cricket club. The lower half of the map contains a mix of Private and
Government Institution and Community categories as well a Public and Private Open Space
(Passive), car parks, and other unused land.
Figure 10. Map of TPU 123 layer from year 1967
The following layer is also of TPU 123 but from year 1985. As shown in Error!
Reference source not found., this layer only has 8 land use categories. The distribution of the
Business and Office category appears very similar to the 1967 layer but includes some of the
land labeled in the Commercial category. Some of the open space categories have changed
included the Statue Square Gardens which was a passive Public Open Space in 1967 and became
a passive Private Open Space. The active Private Open Space of the Cricket club transformed to
56
a passive Private Open Space. Another noteworthy aspect of the 1985 layer is that it includes
some Institution and Community categories that were ferry piers. When viewed with the modern
basemap, it is clear to see how much land has been reclaimed since this survey.
Figure 11. Map of TPU 123 layer from year 1985
The layer for TPU 122 from year 1967 has 12 land use categories and is depicted in
Error! Reference source not found.. The focal point of this map is the major block of
Institution and Community (Government) category while the surrounding land use is heavily
dominated by the Residential (Tenement Buildings) category.
57
Figure 12. Map of TPU 122 layer from year 1967
Error! Reference source not found.Error! Reference source not found. shows the
layer for TPU 122 from year 1985 reflects almost a complete change in categories. This layer
contains 10 land use categories, again with the focal point on a large block of Institution and
Community but this time under the Private subtype. The dominant category around the rest of the
map is also Residential but in 1985 there no longer was a Tenement Building subtype and instead
the subtype of Zone 1 is used.
58
Figure 13. Map of TPU 122 layer from year 1985
The layer for TPU 114 from year 1972 was the most intricate layer as can be seen in
Error! Reference source not found.. This layer consists of 18 land use categories. The upper
part of the map mainly consists of Institution and Community (Government) and Car Park
categories. It is also clear that in 1972 these areas were along the harbor front as some of the land
use was obviously ferry piers. The rest of the map includes all the other categories with a slight
majority of Residential (Tenement Buildings) and Misc (Vacant Building Land) categories. What
makes this map layer stand out is how detailed it is, showing many blocks divided into many
categories with no apparent organizational order.
59
Figure 14. Map of TPU 114 layer from year 1972
The TPU 114 from year 1982 layer, shown in Error! Reference source not found.,
contains 13 categories and does not include the waterfront areas from the 1972 layer. The
northern part of the map is mainly made up of the Business and Office category while the
southern part is mainly Residential (Zone 1). This layer is not as complexly detailed as the 1972
layer but interestingly there is a cluster of land that is mainly Residential (Zone 1) but has cut
outs with the Commercial category.
60
Figure 15. Map of TPU 114 layer from year 1982
4.3 Web Map
All layers of the created spatial data were brought into ArcGIS Online as a web map
before further development into the application. The web map displays the same map layers as
they were in ArcGIS Pro but does not include the other data that was used to create the layers.
The web map is where the final symbology was edited. Since there are 25 unique land use
categories, it was important to find colors for the polygons that could easily be distinguished.
Creating a table with unique RGB values provided better results than simply choosing default
colors from the color palette. The colors are bright and colorful which contrast the grey basemap.
The web map is where the map was reviewed and deemed ready for export to the application.
Error! Reference source not found. shows the web map in ArcGIS Online with all 6 layers.
61
Figure 16. Web Map of all TPU layers in ArcGIS Online
4.4 Web App
When the web app is launched, users arrive to a page with a light grey background and
control panel and in the center of the page is a brightly colored section of the map of Hong Kong
(Error! Reference source not found.). These are the layers of the application that are turned on
by default. Users can immediately start zooming and panning the map with their mouse directly
or use the zoom in and out buttons. Along the bottom of the page is a ribbon in darker grey that
contains the widgets to access the functionality of the application.
62
Figure 17. Landing page of the web application
The widgets open up to smaller windows that stack on top of each other as depicted in
Error! Reference source not found.. All widgets can be moved anywhere on the page except
for the Swipe widget. When the Swipe widget is activated, it splits the page down the line
vertically and minimizes the control ribbon. To exit out of the Swipe widget, the user must click
on the three dots on the minimized control ribbon. At the very bottom of the page there is a small
tab that opens the attribute table. When it is open it consumes one third of the page by default but
can be dragged to take up the entire length of the page.
63
Figure 18. Layout of web app with all widgets engaged
Users can open as many widgets as they like and use the page of the app to spread them
out as shown in Error! Reference source not found.. This is an important feature, as accessing
multiple widgets simultaneously is essential to getting the full functionality from the application.
An academic user could start here to decide how to use the app for their specific analysis needs.
64
Figure 19. Layout of widgets open and moved across page
When using the Swipe widget, the widget toolbar minimizes and the screen is split down
the middle as shown in Error! Reference source not found.. A drop-down menu appears in the
lower right corner where the user can select the layer they want to compare from what is already
displayed on the map. There is a button on the vertical line splitting the screen that when
selected, allows users to view between the different layers by moving the vertical line left or
right. A Planning Department user could use this function as a quick way to observe if a TPU
layer warranted further research on its land use change over time.
65
Figure 20. Example of Swiping widget
The user can access the attribute table for the layers in the app by clicking the small tab at
the bottom center of the page. The default display of the table takes up approximately one third
of the page as shown in Error! Reference source not found. but the user can adjust as needed.
This function could be particularly useful for a GIS student user who is also involved in an HGIS
project.
66
Figure 21. Attribute table opened from bottom of app screen
The Filter widget allows for the user to search and view components of the layers based
on their categories. Error! Reference source not found. shows the example of filtering for the
Business and Office category for the TPU114 1972 layer where only polygons that correspond to
that type are displayed. This widget could be used by a land developer who is interested in a
particular land use category.
67
Figure 22. Filtering Business and Office land use type from layer TPU114 1972
The Screening widget is the widget used for when users want to download shapefiles of a
selection. Using the “draw” function in the widget allows users to draw a box around the area of
interest. Once the selection is made, users can select the download icon and choose their desired
format, an example is shown in Error! Reference source not found.. This widget was added
specifically so that the application could provide open data to users.
68
Figure 23. Screening widget opened to access shapefile download function
The About widget opens to a pane with an introduction to the Hong Kong Historical
Urban Land Use web app as seen in Error! Reference source not found.. Included in the
introduction is a brief statement describing what the project is about, who it can be used by and a
link to download the entire dataset on github. There is also a short step by step guide to
navigating the Screening widget to download specific elements of the dataset.
69
Figure 24. About widget opened with introduction and user tips
The Add Data widget allows users to bring other data into the app to visually analyze it
with the built-in land use layers. Users that have an ArcGIS Online account can search their own
content, their organizations content, or the Living Atlas. Alternatively, if the user does not have
an ArcGIS Online, they can bring in data from a file. Error! Reference source not found.
shows an example of data brought in depicting the original coastline of Hong Kong. Government
Records Office users could use this widget when responding to historical enquiries.
70
Figure 25. Add Data widget opened with original coastline and land reclamation data added
The URL for the web application is https://arcg.is/1GHWPW0. A video tutorial was
made and uploaded to Youtube to show users how to use the app and its functions, the URL to
the tutorial is https://youtu.be/s05ZcU3CumY. In an effort to make the data truly open, the full
dataset of all TPU layers is hosted on a github page for public access and download. The URL is
https://github.com/geoHKG/HK-Historical-Urban-Land-Use.
71
Chapter 5 Discussion
The Hong Kong Historical Urban Land Use Web App has proven to be a successful HGIS
project and is in alignment with the main motivation of creating a user friendly, online database
of historical data. For discussion, this chapter contains three sections. Frist, section 5.1 reviews
the successes of the project. Section 5.2 discusses the potential uses of the application. Section
5.3 discusses an interesting insight about the Tenement Building category gained from using the
application. Finally, Section 5.4 discusses the challenges and limitations of the project as well as
what future work could improve it.
5.1 Project Successes
Overall, the project achieves its intended goals. It is a successful application of an HGIS
project that has utility in various academic and government workflows. It brings a needed
upgrade to data that was previously virtually unusable. Before bringing in this historical data into
to the Hong Kong Historical Urban Land Use Web App, this data would have likely continued to
be unused. Not only does the application improve the usability of this data, it also creates an
alternative method of preservation as paper maps are at risk of deterioration.
Another major success of the project is that is provides a clear and reusable framework
for other HGIS projects. Digitizing historical maps could be considered a must-do for countless
history, GIS, and land/urban planning departments. Eventually, all historical maps may need to
be digitized to be properly preserved for future generations, this project highlights a
straightforward methodology for developing a modern GIS application to address that need.
72
5.2 Utility of the Web App
The application does more than just preserve the historic land use data from older source
maps, it allows for easy visual analysis of how land use of the 1960s-1980s fits into the greater
story of Hong Kong’s urban history. By way of an example, this section now describes an
interesting insight about historical land use classification that was learned by the author while
exploring the Hong Kong Historical Urban Land Use Web App.
The change in the land use classification from Residential (Tenement Buildings) to
Residential (Zone 1) is an obvious land use change detected via the app and although both
categories are Residential, it begs the question as to what happened for the subcategories to
change over the decade time span between the two surveys.
The key to understanding this subcategory land use change is understanding the urban
architectural history of tenement buildings in Hong Kong. The older streets of Hong Kong
maintain the same topography and layout as when Hong Kong was first established. The
prevailing layout being the “shophouse” style of buildings, in which the street-level space was
dedicated to commerce and the upper floors designated for living space. The history of Hong
Kong’s architecture has two defining styles, the colonial-influenced tong lau that dates from pre-
World War II, and the high-rise composite buildings that came after the war.
The literal translation of tong lau is “Chinese house” but the building type was often
referred to as “tenement houses” in British colonial records (Davis 2019). These shophouse
buildings date back to the mid-19th century when they were pioneered by merchants across
South China and Southeast Asia and have since become a critical part of Hong Kong’s urban
landscape. After Hong Kong Island became a British crown colony in 1843, many people from
China’s coastal cities emigrated to supply manual labor for the construction of the city of
73
Victoria. These types of buildings were a response to the needs of the original builders and were
a convenient and economical way of addressing the growing housing problem at the time. The
sociopolitical factors of Hong Kong at that time are what made the local versions of tong lau
unique, but they featured European variations that spoke to their colonial origins. The pre-war
tong lau were between three and four stories. There are not many examples of pre-war tong lau
that remain in their original state. While there have been attempts to restore the tong lau, most of
them have vanished due to Hong Kong’s rapid redevelopment. There are only a few examples of
pre-war tong lau that have been preserved and renovated as an homage to Hong Kong’s urban
heritage. Most of these remodeled tong lau were dilapidated but then converted to community
spaces or upscale flats by wealthy individuals.
After World War II, there was a massive influx of refugees escaping the conflict as well
as the subsequent communist takeover of mainland China. Hong Kong desperately needed to
meet the demands of the exponential rise in population, but the pre-war tong lau was an
insufficient housing solution. They were soon replaced with composite buildings in the 1950s
and 1960s which were up to eight stories tall. According to the Buildings Ordinance they are in
the same vein as the earlier versions, built for both for commercial and domestic purposes (Davis
2019). Essentially, the composite buildings were just an upsized version of the tong lau, with
ground level reserved for retail and the floors above for living spaces, although it was common
for business offices to operate on upper floors and living spaces to be subdivided into smaller
units.
Composite buildings are the hallmark of how Hong Kong began to expand upward on the
z-axis. Hong Kong’s population was increasing by over fifty thousand per year, and the
Buildings Ordinance allowed for construction of up to 9 stories without a lift (Davis 2019).
74
Many developers leapt at the opportunity. This first wave of high-rises had little to no restrictions
on what their internal use could be, so they quickly developed into complicated and crowded
buildings. Between 1959 and 1979, approximately five thousand composite buildings were
constructed, with five hundred of those buildings housing more than fifteen hundred residents
(Davis 2019). Today, many composite buildings are being replaced as there is a government
policy to sell buildings that are over fifty years old.
This history helps paint a more complete picture when understanding the changing of the
land use categories on the TPU maps from the late 1960s to the mid-1980s. Both the tong lau
and the composite buildings would have been classified under “Tenement Buildings” in the
earlier versions of the land use surveys. When compared to the later versions where the
classification is considered “Residential (Zone 1)”, it can be inferred that the physical
construction of the buildings within those polygons have undergone noticeable redevelopment to
meet the growing housing needs. This change in buildings structure, and the development history
it speaks to, is thus made apparent in the Hong Kong Historical Urban Land Use Web App.
5.3 Esri Versus Open Geospatial Software
The choice of using the Esri product suite was made in comparison to the open geospatial
software option. The considerations for the selection were made around functionality, and
scalability and performance.
ArcGIS offers a wide range of advanced functionalities for geospatial analysis and data
visualization. Open geospatial software also has this functionality but not as comprehensive as
ArcGIS. For the purposes of the HK HULU app, both options offer the functionality needed.
Future versions of the application can be expected to include more historical datasets, and
therefore scalability is an important consideration. Open geospatial software can be scalable but
75
could require customization and configuration. Esri software is natively capable of handling
large datasets with high performance. Since one of the core software requirements for the
application is that it must allow for development without any code, Esri was the superior option.
5.4 Challenges/Limitations
There are some limitations to the project that may affect its final output. It is important to
note that this HGIS project does not claim to be entirely accurate. This project relies heavily on
the source maps it was built on. Some of the source maps were difficult to read accurately as the
colors on the paper had faded, making it difficult to discern if a certain area was one category or
another.
The TPU maps included building lines but in order to streamline the digitizing process,
some buildings were lumped together when creating the digital polygons. This inhibits the detail
of analysis that can be done. There was also some uncertainty regarding legibility of the paper
maps’ labels and legend, this could have impacted the accuracy of the land use categories.
Georeferencing was a fundamental part of the methodology for this project.
Georeferencing heavily relies on the GIS operator to make the correct choices of control points.
There are some best practices to increase the accuracy of georectification such as choosing points
in the corners of the source maps and using known historical or geological landmarks that can be
assumed to be consistent from the historical map to the modern one. But the fact remains that
georeferencing is not an exact formula and some degree of inaccuracy is to be expected.
Although the TPU maps were georeferenced to a high standard, minor inaccuracies can be seen
when zooming in on polygons from two layers of the same TPU. In some cases, the border of
polygons from one layer can be seen sticking out slightly compared to the other. Since the source
maps were hand traced in ArcGIS Pro, if one layer is georeferenced slightly different from its
76
corresponding layer of a different temporal scale, then these minor inaccuracies are likely to
occur.
Another minor limitation is regarding the basemap. The choice of basemap was made
based on what would best visualize the overlaid digitized layers. Of the basemaps available in
ArcGIS Pro, the light grey basemap was most appropriate option. However, the references labels
connected to the basemap could not be removed or customized, resulting in some labels still
being displayed when the TPU layers were selected. The accuracy of the web map may be
affected as the labels reference a modern map and may not perfectly align with the historical
layers.
5.5 Future Work
Further work for the future could include including Outline Zoning Maps into the web
app. Outline Zoning maps are official maps that provide a comprehensive framework for land
use planning in Hong Kong. The Hong Kong Planning Department is responsible for maintaining
and updating the Outline Zoning maps. These maps divide the land in Hong Kong into various
zones based on the intended use of that land, whether it be residential, commercial, industrial,
recreational, or agricultural. The zones are labeled to indicate the permitted land use, as well as
any specific development restrictions. Outline Zoning maps are used to guide the development of
land in accordance with Hong Kong’s long-term social, economic, and environmental goals.
At the time of this project’s completion, the application contains six TPU layers. The
Government Records office holds many more land use survey maps of TPUs that encompass all
of Hong Kong Island and Kowloon. A more complete version of this application would include
digitized versions of all the recorded TPU land use surveys. Outline Zoning Plan data is
complementary to land use data as it shows how the land was zoned for its intended use. There
77
are paper maps of historical Outline Zoning Plans available at the Hong Kong Planning
Department’s enquiry desk. The complete version of this application would include all historical
TPU land survey maps and all historical Outline Zoning Plan maps. This would result in a rich
HGIS application that preserves the historical data and is complete enough for deep investigative
analysis.
More functionalities could be added to a future version of the application to make it more
robust. Adding a dashboard that could provide summary statistics of the change in land use and
reorganization of urban space could save time for future users when performing analysis.
Flagging specific areas where there was uncertainty during the digitizing process could increase
the transparency of the application. Finally, adding a time animation slider tool could improve
the visual experience for all users by allowing them to see a quick and clear overview of how the
land use categories changed over time.
78
References
ArcGIS Pro Help. n.d. “Overview of Georeferencing” Accessed March 1, 2023.
https://pro.arcgis.com/en/pro-app/latest/help/data/imagery/overview-of-
georeferencing.htm.
Bai, X, P. Shi, and Y. Liu. 2014. "Society: Realizing China's urban dream." Nature 509, no.
7499. 158-160.
Berman, M. L. 2014. "Boundaries or Networks in Historical GIS: Concepts of Measuring Space
and Administrative Geography in Chinese History." Historical Geography 42. 59-80.
Black, F.A., MacDonald, B.H., & Black, J.M.W. 1998. “Geographic Information Systems: A
New Research Method for book history.” Book History 1, 11-
31. doi:10.1353/bh.1998.0001.
Blatt, H. Tracy, R. and Owens, B. 2006 Petrology: igneous, sedimentary, and metamorphic.
Macmillan.
Bodenhamer, D.J. 2008. "History and GIS: Implications for the Discipline." Placing history:
How maps, spatial data, and GIS are changing historical scholarship. 219-234.
CCICED, Task Force.2013. "Ecosystem issues and policy options addressing the sustainable
development of China’s ocean and coasts." China Environmental Science Press,
Beijing 493
Charalabidis, Y., A. Zuiderwijk, C. Alexopoulos, M. Janssen, T. Lampoltshammer, and E. Ferro.
2018. “The World of Open Data.” Public Administration and Information Technology.
978-3.
Chen, R. and Xie, J. 2008. "Open-Source Databases and Their Spatial Extensions." Open-
Source Approaches in Spatial Data Handling. 105-124.
Chignard, S. March 29, 2013. “A Brief History of Open Data.” ParisTech Review,
https://www.paristechreview.com/2013/03/29/brief-history-open-
data/#:~:text=Open%20data%20responds%20to%20a,of%20geophysical%20and%20env
ironmental%20data.
Cillis, G., D. Statuto, P. Picuno. 2021. "Historical GIS as a Tool for Monitoring, Preserving and
Planning Forest Landscape: A Case Study in a Mediterranean Region." Land 10, no. 8.
851.
Civatis. n.d. “History of Hong Kong - Past, Present and Future.” Introducing Hong Kong .
Accessed February 20, 2023. https://www.introducinghongkong.com/history
79
Civil Engineering & Development Department (CEDD). June, 30, 2019. “The geology of Hong
Kong” Introduction. https://www.cedd.gov.hk/eng/about-
us/organisation/geo/pub_info/memoirs/geology/int/index.html.
Coetzee, S., I. Ivánová, H. Mitasova, and M. A. Brovelli. 2020. "Open Geospatial Software and
Data: A Review of the Current State and A Perspective into the Future." ISPRS
International Journal of Geo-Information 9, no. 2. 96.
Cunfer, G. 2005. On the Great Plains: Agriculture and Environment. No. 20. Texas A&M
University Press
Data.gov.hk. n.d.“FAQ.” Accessed March 5, 2023. https://data.gov.hk/en/faq.
Davis, Hana. “How Hong Kong’s Historic Tong Lau and Composite Buildings Shaped the City’s
Built Environment.” South China Morning Post, June 15, 2019.
https://www.scmp.com/news/hong-kong/health-environment/article/3014526/upstairs-
downstairs-how-hong-kongs-historic-
tong?module=perpetual_scroll_0&pgtype=article&campaign=3014526.
Ellul, C., N. Tamash, F. Xian, J. Stuiver, and P. Rickles. 2013 "Using Free and Open Source GIS
to Automatically Create Standards-Based Spatial Metadata in Academia-First
Investigations." In Free and Open Source Software for Geospatial (FOSS4G) Conference
Proceedings, vol. 13, no. 1, p. 8.
Esri. 1994. ArcInfo Data Management: Concepts, Data Models, Database Design, and Storage.
Redlands, CA: Environmental Systems Research Institute.
Fu, Y. B., K. Cao, F. Wang, and F. S. Zhang. 2010. "Primary quantitative evaluation method on
sea enclosing and land reclamation strength and potential." Ocean Development and
Management 1, no. 27. e30.
Gregory, I. N., and P. S. Ell. 2007. Historical GIS: Technologies, Methodologies, and
Scholarship. Cambridge Studies in Historical Geography. Cambridge: Cambridge
University Press. DOI: 10.1017/CBO9780511493645.
Gregory, I. N., and R. G. Healey. 2007. "Historical GIS: structuring, mapping and analysing
geographies of the past." Progress in Human Geography 31, no. 5. 638-653
Gregory, W. G. 1964. "An architect’s comments on land use in Hong Kong." Land use problems
in Hong Kong. 20-39.
Hu, J. 2012. "Full text of Hu Jintao’s report at 17th Party Congress." Xinhua News. Consultado
el 20. 2007-10.
Johnson, P. A., R. Sieber, T. Scassa, M. Stephens, and P. Robinson. 2017. "The cost (s) of
geospatial open data." Transactions in GIS 21, no. 3. 434-445.
80
Klemme, A.. 2014. "International Dunhuang Project: The Silk Road Online." Reference Reviews
28, no. 2. 51-52. doi:10.1108/RR-09-2013-0224.
Knowles, A. K., and A. Hillier, eds. 2008. Placing history: how maps, spatial data, and GIS are
changing historical scholarship. Redlands, CA: Esri Pres.
Kwok, S. 2012. “Geodetic Datums and Coordinate Systems for Surveying and Mapping in Hong
Kong,”
https://www.fig.net/resources/proceedings/fig_proceedings/fig2012/ppt/ts02b/ts02b_kwo
k_6113_ppt.pdf.
Lancaster, L. R., and D. J. Bodenhamer. 2002. The electronic cultural atlas initiative and the
North American religion atlas. na.
Lau, S. S. Y., Renganathan Giridharan, and Sankar Ganesan. 2005. "Multiple and intensive land
use: Case studies in Hong Kong." Habitat International 29, no. 3. 527-546.
Lawson, K., R. Bavaj, and B. Struck. 2022. “A Guide to Spatial History: Areas, Aspects, and
Avenues of Research.” Olsokhagen.
Li, H, Y. D. Wei, F. H. Liao, and Z. Huang. 2015. "Administrative hierarchy and urban land
expansion in transitional China." Applied Geography 56. 177-186.
Liu, W, and B. Q. Liu. 2008. "Current situation and countermeasures of sea reclamation in
China." Guangzhou Enviromental Sciences 23, no. 02. 26-30.
Maurya, S. P., A. Ohri, and S. Mishra. 2015. "Open-Source GIS: A Review." In Proceedings of
National Conference on Open-Source GIS: Opportunities and Challenges, Department of
Civil Engineering, IIT (BHU), Varanasi, 150. ISBN: 978-81-931-2500-7.
McLeman, R., Herold, S., Reljic, Z., Sawada, M., and McKenney, D. 2011. "GIS-based
modeling of drought and historical population change on the Canadian Prairies."
Population and Environment 33, no. 2-3. 161-185.
Minghini, M., A. Mobasheri, V. Rautenbach, and M. A. Brovelli. 2021. "Geospatial Openness:
From Software to Standards & Data." ISPRS International Journal of Geo-Information
10, no. 12. 753.
Neteler, M., M. H. Bowman, M. Landa, and M. Metz. 2012. "GRASS GIS: A multi-purpose
open source GIS." Environmental Modelling & Software 3. 124-130.
Ng, Justin. 2022. “The Importance of Land Reclamation in Hong Kong and Its Impacts.”
Earth.Org, https://earth.org/land-reclamation-hong-kong/.
Open Data Handbook. n.d. “What is open data?” Accessed March 5, 2023.
https://opendatahandbook.org/guide/en/what-is-open-data/.
81
Open Knowledge Foundation. n.d. “What is open?” Accessed February 3, 2023.
https://okfn.org/opendata/.
Ostrom, E. 1990. “Governing the commons: The evolution of institutions for collective action.”
Cambridge University Press.
Oxford, S., D. Dorling, R. Mitchell, M. Shaw, and G. D. Smith. 2001. "Life and death of the
people of London: a historical GIS of Charles Booth's inquiry." Health & Place 7, no. 1.
27-46.
Planning Department. 2018 “About Us” Accessed February 3, 2023.
https://www.pland.gov.hk/pland_en/about_us
–––– . 2021. “About Hong Kong 2030+.” Hong Kong 2030+. Accessed March 20, 2023.
https://www.pland.gov.hk/pland_en/p_study/comp_s/hk2030plus/about_a.htm#:~:text=%
E2%80%9CHong%20Kong%202030%2B%E2%80%9D%20is,of%20Hong%20Kong%2
0beyond%202030.
Raifer, M., R. Troilo, F. Kowatsch, M. Auer, L. Loos, S. Marx, K. Przybill, S. Fendrich, F. B.
Mocnik, and A. Zipf. 2019. "OSHDB: a framework for spatio-temporal analysis of
OpenStreetMap history data." Open Geospatial Data, Software and Standards 4. 1-12
Re Calegari, G., E. Carlino, D. Peroni, and I. Celino. 2015. "Extracting urban land use from
linked open geospatial data." ISPRS International Journal of Geo-Information 4, no. 4
2109-2130.
Schaefer, M. 2004. "Design and Implementation of a Proposed Standard for Digital Storage and
Internet-based retrieval of Data from the Tithe Survey of England and Wales." Historical
Methods: A Journal of Quantitative and Interdisciplinary History 37, no. 2. 61-72.
Schenk, C.R. n.d. “Economic History of Hong Kong.” EHnet. University of Glasgow. Accessed
March 20, 2023. https://eh.net/encyclopedia/economic-history-of-hong-
kong/#:~:text=The%20industrialization%20of%20Hong%20Kong,the%20embargoes%2
0of%20the%201950s.&text=Hong%20Kong%27s%20industry%20was%20founded,inte
nsive%20production%20mainly%20for%20export.
Schlichting, K..2008. "Historical GIS: New Ways of Doing History." Historical Methods: A
Journal of Quantitative and Interdisciplinary History 41, no. 4. 191-196.
Doi:10.3200/HTMS.41.4.191-196.
Sejati, A. W., I. Buchori, I. Rudiarto, C. Silver, and K. Sulistyo. 2020. “Open-Source Web GIS
Framework in Monitoring Urban Land Use Planning: Participatory Solutions for
Developing Countries” Journal of Urban & Regional Analysis 12, no. 1
Sengupta, D, R. Chen, and M. E. Meadows. 2018. "Building beyond land: An overview of
coastal land reclamation in 16 global megacities." Applied Geography 90. 229-238.
82
Shaw, R, D. L. K. Tang, R. B. Owen, and R. J. Sewell. 2010. "The geological history of Hong
Kong." Asian Geographer 27, no. 1-2. 43-57.
Siebert, L. 2000. "Using GIS to document, visualize, and interpret Tokyo’s spatial history."
Social Science History 24, no. 3. 537-574.
Skopyk, B. March 8, 2021.“Georeferencing Historical Maps.” ArcGIS StoryMaps. Esri,
https://storymaps.arcgis.com/stories/dd75d0398f7d4ded924d303161895b8b.
Statuto, D., G. Cillis, and P. Picuno. 2017. "Using Historical Maps within a GIS to Analyze Two
Centuries of Rural Landscape Changes in Southern Italy." ISPRS International Journal of
Geo-Information 6, no. 9. 274.
Steiniger, S., and E. Bocher. 2009. "An overview on current free and open source desktop GIS
developments." International Journal of Geographical Information Science 23, no. 10.
1345-1370.
Sui, D. 2014. "Opportunities and impediments for open GIS." Transactions in GIS 18, no. 1. 1-
24
Szczepanski, K. June 9, 2020. “Why Did Hong Kong Belong to Britain?” ThoughtCo.
ThoughtCo, https://www.thoughtco.com/china-lease-hong-kong-to-britain-195153.
The International Dunhuang Project. n.d. “The Silk Road Online.” Accessed January 20, 2023.
http://idp.bl.uk/.
The World Bank. 2019. “Open Data Essentials.”
http://opendatatoolkit.worldbank.org/en/essentials.html
Tian, B, W. Wu, Z. Yang, and Y. Zhou. 2016. "Drivers, trends, and potential impacts of long-
term coastal reclamation in China from 1985 to 2010." Estuarine, Coastal and Shelf
Science 170. 83-90
Town Planning Board. 2022.“About Us.” Introduction, Accessed March 1, 2023.
https://www.info.gov.hk/tpb/en/about_us/intro.html.
Vetter, David. July 28, 2018. “Hong Kong Land Reclamation Explained: The Good, Bad and
Ugly Methods of Pushing Back the Sea.” South China Morning Post.
https://www.scmp.com/news/hong-kong/community/article/2156638/hong-kong-land-
reclamation-explained-good-bad-and-ugly.
Wang, H, X. Zhang, and M. Skitmore. 2015. "Implications for sustainable land use in high-
density cities: Evidence from Hong Kong." Habitat international 50. 23-34.
Wang, W, H. Liu, Y. Li, and J. Su. 2014. "Development and management of land reclamation in
China." Ocean & Coastal Management 102. 415-425.
83
Wilkinson, M. D., M. Dumontier, I. J. Aalbersberg, G. Appleton, M. Axton, A. Baak, N.
Blomberg et al. 2016. "The FAIR Guiding Principles for scientific data management and
stewardship." Scientific Data 3, no. 1. 1-9.
Wilson, A. 2001. "Sydney timemap: Integrating historical resources using GIS." History and
Computing 13, no. 1. 45-69.
Wong, Y. G. 2011. "The puzzle and thinking on Hong Kong’s little land but large
population." Ming Bao Daily News. December 13, 2011.
Ye, S. 1998. "The land resource and landuse of Hong Kong." Chinese Geographical Science 8,
no. 1. 12-24.
Abstract (if available)
Abstract
Hong Kong is a dynamic city located on the southern coast of mainland China. A once unassuming island of fishing villages became an economical trading hub as a British crown colony following the Opium Wars. Hong Kong’s geology limited the natural area of developable land, and as the population of the colony increased over the decades, land reclamation projects were commissioned to account for exponential emigration. Over the course of its century-long history, the urban topography of Hong Kong has transformed significantly, and is expected to continue to evolve as the city maintains its status as an international business hub. This thesis explores the development of an open Historical Geographic Information System (HGIS) web application that portrays interactive digital versions of historic land use maps of Hong Kong. Historic maps of Tertiary Planning Units in Hong Kong’s Central district from the late 1960s to the late 1980s are the maps used in this application’s first iteration. This project incorporates georeferencing and spatial data creation techniques and methodologies for digitizing historical data and configuring app building software. The Hong Kong Historic Urban Land Use Web App successfully delivers the historic data in a user-friendly web-based application that allows users to investigate the land use differences and download the complete dataset for their individual purposes. The project also serves as a model for the development of similar web-based applications for exploring historical spatial data.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Advancing Redwood City's bicycle infrastructure through a geodesign workflow
PDF
The community wildfire protection plan repository: using VGI to create a national collection of wildfire management information
PDF
Maui’s wildland-urban interface: enhancements for the unique vegetative and agricultural landscape
PDF
Exploring urban change using historical maps: the industrialization of Long Island City (LIC), New York
PDF
Historical observations of wildlife in Kenya: a Web GIS application
PDF
Recreational Off-road Adventure Motorcycle mapping System (ROAMS): a web application facilitating adventure motorcycling in Idaho public lands
PDF
Exploring the pernicious effects of redlining and discriminatory policies on an American city: a spatio-temporal case study of New York City
PDF
Human suffering during wartime: a StoryMap of violations of international law during the Russo-Ukrainian War
PDF
Rails to trails web mapping application for the great Redwood trails: mapping northern California’s repurposed trails
PDF
Trojan Food Finder: a web-based GIS campus food sharing application
PDF
Development of a Web GIS for urban sustainability indicators of Oakland, California
PDF
A spatial analysis of veteran healthcare accessibility
PDF
Creating a well database and web mapping application: using geographic information systems to manage and monitor groundwater resources in Sonoma County, California
PDF
Generating bicyclist counts using volunteered and professional geographic information through a mobile application
PDF
That sinking feeling: predicting land subsidence in California’s San Joaquin Valley with a spatial regression model
PDF
Assessing the reliability of the 1760 British geographical survey of the St. Lawrence River Valley
PDF
Spatiotemporal analysis of the SLOSH and ADCIRC storm surge models: a case study of hurricane Ida
PDF
Generating trail conditions using user contributed data through a web application
PDF
Designing an earthquake preparedness web mapping application for the older adult population of Los Angeles, California
PDF
Assessing the impact of a web-based GIS application to promote earthquake preparation on the University of Southern California University Park Campus
Asset Metadata
Creator
Patton, Brandon Noel
(author)
Core Title
Development of a historical urban land use web application for the city of Hong Kong
School
College of Letters, Arts and Sciences
Degree
Master of Science
Degree Program
Geographic Information Science and Technology
Degree Conferral Date
2023-08
Publication Date
06/02/2023
Defense Date
05/11/2023
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
GIS,Historical,Hong Kong,Land use,OAI-PMH Harvest,web application
Format
theses
(aat)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Sedano, Elisabeth J. (
committee chair
), Huang, Guoping (
committee member
), Vos, Robert O. (
committee member
)
Creator Email
bpatton@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-oUC113147822
Unique identifier
UC113147822
Identifier
etd-PattonBran-11920.pdf (filename)
Legacy Identifier
etd-PattonBran-11920
Document Type
Thesis
Format
theses (aat)
Rights
Patton, Brandon Noel
Internet Media Type
application/pdf
Type
texts
Source
20230606-usctheses-batch-1052
(batch),
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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.
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
Repository Email
cisadmin@lib.usc.edu
Tags
GIS
web application