University of Southern California Dissertations and Theses

Historical observations of wildlife in Kenya: a Web GIS application

(USC Thesis Other)

Historical observations of wildlife in Kenya: a Web GIS application

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content HISTORICAL OBSERVATIONS OF WILDLIFE IN KENYA:
A WEB GIS APPLICATION
by
Tianle Lin
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)
May 2025
Copyright 2025 Tianle Lin

ii
To Mom and Dad

iii
Acknowledgements
I would like to thank Dr. Ruddell, Dr. Loyola, Dr. Swift, and Dr. Osborne for their precious
guidance, support and encouragement throughout this project. Their expertise and feedback have
been instrumental in shaping the direction and quality of my work. I appreciate all my incredible
colleagues and classmates, who have consistently supported me on and shown understanding
throughout this challenging yet rewarding journey. Finally, I am very grateful to my friends and
family, who have been my constant source of comfort and inspiration. Your support means the
world to me.

iv
Table of Contents
Dedication.......................................................................................................................................ii
Acknowledgements........................................................................................................................iii
List of Tables ................................................................................................................................. vi
List of Figures............................................................................................................................... vii
List of Abbreviations ...................................................................................................................... x
Abstract.......................................................................................................................................... xi
Chapter 1 Introduction .................................................................................................................... 1
1.1 Motivation........................................................................................................................... 2
1.2 Study Area .......................................................................................................................... 5
1.3 Methods Overview.............................................................................................................. 6
1.3.1 Data Collection and Processing ................................................................................ 6
1.3.2 Data Visualization ..................................................................................................... 6
1.3.3 Web Map Development.............................................................................................. 7
1.3.4 New Observation Report............................................................................................ 7
1.3.5 Web Appearance Design............................................................................................ 8
1.4 Thesis Overview ................................................................................................................. 8
Chapter 2. Related Work............................................................................................................... 10
2.1 GIS in Animal Conservation............................................................................................. 10
2.1.1 GIS Mapping of Subtropical Estuarine Habitats..................................................... 10
2.1.2 Integrating GIS and Remote Sensing for Wildlife Management.............................. 12
2.1.3 GIS in Wildlife Tracking .......................................................................................... 14
2.2 Web Map Design for Animal Conservation ............................................................... 15
2.2.1 A New Way to Map with Conservation Data.................................................. 15
2.2.2 Advanced Species Mapping for Conservation ................................................ 16
2.3 Volunteered Geographic Information in Wildlife Conversation ................................ 18
2.3.1 Uncertainty-Aware Enrichment of Animal Movement Trajectories by VGI... 18
2.3.2 VGI and the Spatial-Discursive Construction of Nature ................................ 19
2.3.3 Assuring the Quality of Volunteered Geographic Information ...................... 21
Chapter 3: Methods....................................................................................................................... 23
3.1 Data Collection and Processing ........................................................................................ 23
3.1.1 Data Source: iNaturalist.......................................................................................... 23

v
3.1.2 Data Extraction........................................................................................................ 27
3.2 Geographic Information System (GIS) Integration .......................................................... 30
3.3 Experience Builder............................................................................................................ 34
3.3.1 Data Export to ArcGIS Online................................................................................. 35
3.3.2 Web Map Design...................................................................................................... 36
3.3.3 Integration in Experience Builder ........................................................................... 43
3.4 New Observation Report................................................................................................... 46
3.5 Web Framework................................................................................................................ 50
3.5.1 Welcome Message Pop-up ....................................................................................... 50
3.5.2 Header, Additional Resources Bar and Help Icon .................................................. 54
3.5.3 Container................................................................................................................. 59
3.5.4 Footer....................................................................................................................... 64
3.5.5 Instruction Video Pop-up......................................................................................... 66
Chapter 4. Results......................................................................................................................... 71
4.1 Interactive Map ................................................................................................................. 71
4.2 Web Appearance............................................................................................................... 78
4.3 Data Collection Survey ..................................................................................................... 83
Chapter 5. Conclusions................................................................................................................. 87
5.1 App Summary ................................................................................................................... 87
5.2 Historical Kenya Wildlife Observation Web Application VS. iNaturalist ....................... 89
5.2.1 Data Focus............................................................................................................... 89
5.2.2 Spatial Analysis........................................................................................................ 89
5.2.3 Map Accessibility ..................................................................................................... 89
5.3 Challenges and Limitations............................................................................................... 90
5.3.1 Data Integration and Cleaning................................................................................ 90
5.3.2 Software Limitations in ArcGIS Suite ...................................................................... 91
5.3.3 User Accessibility and Interface Design.................................................................. 92
5.3.4 Potential Competitors.............................................................................................. 93
5.4 Future Work...................................................................................................................... 93
5.4.1 Enhanced Data Integration...................................................................................... 93
5.4.2 Advanced Data Analytics and Visualization............................................................ 94
5.4.3 Improved User Experience and Interface Design.................................................... 94
5.4.4 Enhanced Data Security and Privacy ...................................................................... 95
5.4.5 Expanded Educational and Outreach Features....................................................... 95
5.4.6 Broader Geographic Expansion .............................................................................. 96
References..................................................................................................................................... 97

vi
List of Tables
Table 1. SWOT Analysis for iNaturalist....................................................................................... 26
Table 2. Potential Solutions to iNaturalist’s Weakness and Threat.............................................. 27
Table 3. Data Types and Purposes in the Project.......................................................................... 29

vii
List of Figures
Figure 1. Study Area, the Kenya National Boundary ..................................................................... 5
Figure 2. XY Table to Points Tool................................................................................................ 31
Figure 3 ArcGIS Online Layer Management................................................................................ 36
Figure 4. Basemap Editing............................................................................................................ 37
Figure 5. Adding Map Contents.................................................................................................... 38
Figure 6. Data Clustering Setting.................................................................................................. 39
Figure 7. Pop-ups Setting.............................................................................................................. 40
Figure 8. Example of a Pop-up Window ...................................................................................... 41
Figure 9. Time Slider Setting........................................................................................................ 42
Figure 10. Experience Builder Map Appearance.......................................................................... 43
Figure 11. Map Content Appearance on Small Screen Device .................................................... 44
Figure 12. Search Box Design ...................................................................................................... 45
Figure 13. Esri Survey123 Form Appearance............................................................................... 47
Figure 14. Esri Survey123 Creation.............................................................................................. 48
Figure 15. Welcome Message Pop-up .......................................................................................... 51
Figure 16. HTML Expression for the Pop-Up.............................................................................. 52
Figure 17. JavaScript Expression for the Pop-up Modal .............................................................. 52
Figure 18. CSS Expression for the Welcome Message ................................................................ 53

viii
Figure 19. Header Appearance ..................................................................................................... 54
Figure 20. HTML Expression for the Header and Part of Elements in the Container.................. 55
Figure 21. CSS Expression for the Header................................................................................... 56
Figure 22. CSS Expression for Additional Resources Bar ........................................................... 57
Figure 23. CSS Expression for the Help Icon............................................................................... 58
Figure 24. Maps and Other Elements in the Container................................................................. 59
Figure 25. HTML for the Container and “New Observation” Button .......................................... 60
Figure 26. CSS Expression for the Container Structure and Layout ............................................ 61
Figure 27. CSS Expression for iframe .......................................................................................... 63
Figure 28. CSS Expression for the New Observation Link .......................................................... 64
Figure 29. Footer Appearance....................................................................................................... 64
Figure 30. HTML Expression for the footer................................................................................. 65
Figure 31. CSS Expression for the Footer .................................................................................... 65
Figure 32. Hovering Text for Help Icon ....................................................................................... 66
Figure 33. Instruction Video Appearance..................................................................................... 66
Figure 34. HTML Expression for the Instruction Video .............................................................. 67
Figure 35. CSS Expression for the Instruction Video................................................................... 68
Figure 36. Historical Kenya Wildlife Observation App ............................................................... 70
Figure 37. Web Map Appearance ................................................................................................. 72
Figure 38. A Closer look for Symbology of Data Points.............................................................. 73

ix
Figure 39. Layers Visualization.................................................................................................... 74
Figure 40. Basemap Options......................................................................................................... 75
Figure 41. Pop-up Example .......................................................................................................... 76
Figure 42. Hotspot Analysis for Amphibia................................................................................... 77
Figure 43. Web Appearance ......................................................................................................... 78
Figure 44. Wildlife Observation App Mobile View ..................................................................... 80
Figure 45. Welcome Message....................................................................................................... 81
Figure 46. Instruction Video ......................................................................................................... 82
Figure 47. New Observation Form ............................................................................................... 84
Figure 48. Layer Contains New Observed Points (Not Open to Public)...................................... 85

x
List of Abbreviations
API Application Programming Interface
CSV Comma-Separated Values
CSS Cascading Style Sheets
GBIF Global Biodiversity Information Facility
GIS Geographic Information System
GPS Global Positioning System
HTML Hyper Text Markup Language
JSON JavaScript Object Notation
NDVI Normalized Difference Vegetation Index
RS Remote Sensing
SWOT Strengths, Weaknesses, Opportunities, Threats
UAV Unmanned Aerial Vehicle
VGI Volunteered Geographic Information
WGS 84 World Geodetic System 1984

xi
Abstract
The conservation of wildlife is a global imperative, driven by the need to preserve biodiversity
and maintain the delicate balance of ecosystems. In Africa, a continent renowned for its rich and
diverse wildlife, these efforts are particularly critical. However, the challenges posed by habitat
loss, climate change, and illegal poaching have intensified, necessitating innovative approaches
to wildlife conservation. In response to these challenges, this thesis presents the development and
implementation of the Historical Kenya Wildlife Observation App, a Geographic Information
System (GIS)-based platform designed to document historical and report new observations of
wildlife across Kenya.
The Historical Kenya Wildlife Observation App is an advanced tool that leverages the
power of GIS and community engagement to create a comprehensive database of wildlife
movements and populations. Built on a foundation of user-generated data, sourced primarily
from iNaturalist (iNaturalist 2024), the app provides historical insights into the distribution of
species, helping conservationists, researchers, and policymakers make informed decisions. By
integrating automated data processing, interactive maps, and educational resources, the app
serves as both a scientific tool and a platform for public participation in conservation efforts.
Results of this thesis show the technical development of the Historical Kenya Wildlife
Observation App. This platform integrated GIS and VGI to track historical wildlife observations
in Kenya, cooperated with user interface design techniques and data management strategies. It

xii
also examines the application's potential impact on wildlife conservation, highlighting the role of
digital platforms in fostering community engagement and improvements may applied in the
future.

1
Chapter 1 Introduction
This project is an innovative GIS initiative to enhance the conservation efforts for Kenya's
diverse wildlife. This project utilizes historical data from iNaturalist, focusing on four taxa—
amphibians, mammals, aves, and reptiles—collected between 2010 and 2020. By leveraging
Volunteer Geographic Information (VGI), the project seeks to provide a comprehensive and
interactive platform that enables users to explore historical wildlife observations in Kenya.
A primary objective of this project is to create a powerful, user-friendly, multi-platform
web application that can be accessed and utilized by individuals with varying levels of map-related
knowledge. The application is designed to serve as an informative tool that offers detailed spatial
information on wildlife, including the location, time of observation, and associated images. The
emphasis on historical data ensures that the platform is a repository of past wildlife records and a
tool for understanding long-term trends in species distribution.
In addition to its historical focus, the project includes a feature for reporting new wildlife
observations. This new observation report function allows users to contribute fresh data. These
contributions will be integrated into the public-facing application every five years to reduce the
risk of poaching. This approach balances the need for updated information with the imperative of
wildlife protection.

2
The project also aims to increase public awareness and engagement in wildlife
conservation. By linking to various informative websites, including those dedicated to wildlife
services and anti-poaching efforts, the web map is a gateway for users to explore Kenya's wildlife
and the importance of protecting it. Through this multifaceted approach, the project addresses the
weaknesses and threats (as mentioned in chapter 3.1) associated with existing VGI platforms like
iNaturalist. It builds on their strengths to create a more robust and secure tool for wildlife
conservation in Kenya.
1.1 Motivation
The motivation behind the historical wildlife observation project stems from a confluence
of factors, including the urgent need for effective wildlife conservation strategies, the increasing
availability and potential of VGI, and the limitations of traditional methods of animal observation
and monitoring.
From a conservation perspective, the rapid decline of biodiversity globally (Butchart,
2010), and specifically in regions like Kenya, necessitates innovative approaches to wildlife
monitoring. Traditional methods, such as field surveys and camera traps, while valuable, often
require significant resources, are time-consuming, and are limited in spatial and temporal scope.
These methods, though accurate, can only cover a fraction of the vast and varied landscapes where
wildlife exists, leading to gaps in data and potential oversight of critical changes in wildlife
populations and distributions (Liu, 2024).

3
This project leverages the power of VGI, specifically data sourced from iNaturalist, to fill
these gaps by providing a continuous stream of observations collected by a diverse range of
contributors. Unlike traditional animal documentation methods, which are often restricted to
professional researchers and conservationists, VGI democratizes data collection by enabling
anyone with a smartphone and a passion for nature to contribute valuable information. This
crowdsourcing approach not only increases the volume of data collected but also enhances the
geographic and temporal coverage, offering a more comprehensive view of wildlife distribution
and trends (Siriba, 2017).
Moreover, this project addresses one of the critical challenges associated with VGI—data
accuracy and the potential for misuse. By focusing on historical data (2010-2020) and
implementing a five-year delay on the public release of new observation data, the project mitigates
the risk of poaching, which is a significant concern in regions rich in biodiversity like Kenya. This
approach contrasts with traditional methods, where data might be released immediately, potentially
compromising the safety of endangered species.
Beyond its contributions to conservation and data collection, this project aims to increase
public awareness and engagement in wildlife protection. The user-friendly, interactive web
application provides an accessible platform for individuals to explore and learn about Kenya's
wildlife, encouraging a sense of responsibility and stewardship among a broader audience.
Research has shown that educational engagement with wildlife fosters conservation behaviors and
a deeper connection to biodiversity (Ballantyne & Packer, 2009). Unlike traditional wildlife

4
monitoring methods, which are typically confined to experts, this project actively involves the
public, fostering inclusivity and enhancing collective efforts in conservation.
Ultimately, the motivation for this project is rooted in the desire to create a more effective,
inclusive, and secure method for wildlife monitoring that leverages modern technology and
community participation to address the limitations of traditional animal observation methods. This
project represents a step forward in the use of GIS and VGI for conservation, aiming to not only
protect wildlife but also to empower people to play an active role in the preservation of Kenya's
natural heritage.
In June 2024, midway through the development of this project, Esri and iNaturalist
launched a beta version of a wildlife tracking map. This new tool reflects the increasing
recognition of GIS and VGI as essential biodiversity monitoring and conservation technologies.
While this beta version aligns with the broader goals of this thesis, it was released after the
project’s design and implementation began in January 2024. The existence of such advancements
underscores the relevance of this research. It highlights the importance of historical data
integration and visualization.

5
1.2 Study Area
Kenya is home to a wide range of habitats, from savannas and grasslands to forests,
wetlands, and coastal regions, each supporting a unique array of species. Kenya’s wildlife
includes iconic species such as elephants, lions, giraffes, and rhinoceroses, alongside a rich
variety of bird species, amphibians, and reptiles (Kenya Wildlife Tours, 2025). The map in
Figure 1 provides a detailed overview of Kenya’s geographical and ecological features, focusing
on regions critical to wildlife conservation. The major cities in Kenya are labeled with red points.
The polygons of other crucial wildlife areas including wetlands, national parks, national reserves,
and nation sanctuaries are highlighted with distinct colors.
Figure 1. Study Area, the Kenya National Boundary

6
1.3 Methods Overview
This section outlines the key steps and techniques used to develop the Historical
Observation of VGI Data Web Map for Wildlife in Kenya. The methods employed encompass
several phases, including data collection, data processing and cleaning, GIS integration, web
frame design, and the design and implementation of the ArcGIS Experience Builder for user
interaction.
1.3.1 Data Collection and Processing
The primary data source for this project is iNaturalist, a platform aggregating VGI
contributed by users globally. The project focuses explicitly on historical data from 2010 to
2020, covering four key taxa: Amphibians, Mammals, Aves, and Reptiles within Kenya. The raw
data was extracted from iNaturalist, cleaned, and pre-processed to ensure accuracy and
relevance. This step involved filtering the data to remove duplicates, correcting spatial
inaccuracies, and categorizing the observations according to the selected taxa.
1.3.2 Data Visualization
The processed data was then integrated into a GIS platform to create a spatially accurate
and interactive web map. This process involved georeferencing the observation data to align it
with Kenya's existing geographic frameworks. ArcGIS Pro tools were used for data

7
visualization, classification, and analysis. The map layers were designed to be informative and
user-friendly, enabling users to interact with the data and visually explore wildlife trends.
1.3.3 Web Map Development
The core of this project is the development of an interactive web map designed to be
both powerful and accessible to users with varying levels of technical expertise. The map was
developed using modern web mapping technologies such as Esri ArcGIS products, which has
features that allow users to view and explore wildlife data spatially. The interface includes tools
for zooming, panning, and querying the data, making it easy for users to access detailed
information about each observation, including location, time, and accompanying images.
1.3.4 New Observation Report
A new observation report feature was developed to enhance the functionality of the web
map. This feature allows users to submit new wildlife observations directly through the web
application. These reports are stored in a database and scheduled for public release every five
years to protect sensitive wildlife information from potential misuse, such as poaching. The
observation report system is integrated with Esri Survey123 (Esri 2025), enabling efficient data
collection and management.

8
1.3.5 Web Appearance Design
The web page design focused on user experience, accessibility, and responsiveness. The
website was designed using HTML, CSS, and JavaScript to make it visually appealing and
functional across different devices and screen sizes. The design process involved creating a clean
and intuitive layout that prioritizes ease of navigation while maintaining the integrity of the data
presented. Effective UX/UI principles are critical for web GIS design to ensure accessibility,
usability, and engagement for diverse users." (Nielsen and Norman, 2020)
1.4 Thesis Overview
The remainder of this thesis is structured into four key chapters. Chapter 2: Related
Literature explores relevant topics including the role of VGI in wildlife conservation, its
application in tracking species distribution, and challenges related to the quality and accuracy of
VGI data. The chapter also discusses the strengths and limitations of using GIS for wildlife
monitoring, drawing from prior research. Chapter 3: Methods details the methodological
approach employed in the study, focusing on data collection, cleaning, GIS integration, and the
design of the ArcGIS web application. It also outlines the process of developing the new wildlife
observation reporting system. Chapter 4: Results presents the outcomes of the methodological
steps, including the creation of an interactive web map and the integration of user-submitted
wildlife observations. Lastly, Chapter 5: Conclusions reflects on the findings, discusses the

9
implications for wildlife conservation, and suggests future research avenues to enhance the use
of VGI in environmental protection efforts.

10
Chapter 2. Related Work
This chapter explores the foundational role of emerging technologies in wildlife conservation,
emphasizing the integration of GIS, web map design, and VGI. These tools empower
conservationists and the public to make informed decisions and contribute to effective wildlife
management. This section reviews relevant studies to contextualize the integration of these
technologies, highlighting their potential and challenges that could affect the outcome of the
project.
2.1 GIS in Animal Conservation
GIS play an instrumental role in animal conservation by enabling the mapping and analysis of
migration patterns. By providing spatial data and visualization, conservationists can make datadriven decisions about habitat preservation and design effective wildlife corridors. Additionally,
GIS aids in monitoring human activities like poaching, contributing to more proactive
conservation strategies.
2.1.1 GIS Mapping of Subtropical Estuarine Habitats
Zharikov (2005) demonstrates the application of aerial photography and GIS in the
mapping and characterization of estuarine landscapes, the environments where freshwater from
rivers mixes with saltwater from the sea, creating rich habitats for diverse aquatic species. The

11
study examines the intricate relationships between estuarine landscapes and wildlife by
leveraging aerial photography and GIS to analyze spatial patterns and habitat structures. The
author emphasizes that estuarine environments provide critical ecosystem services and habitats
for diverse wildlife species. Understanding their structure and distribution is crucial for
conservation strategies. The methodology involves capturing high-resolution aerial imagery,
which provides detailed insights into the landscape structure. This imagery is integrated into GIS
to create detailed maps highlighting various landforms, vegetation patterns, and other significant
features. Zharikov (2005) categorizes different habitat types based on their spatial characteristics,
vegetation, and hydrological properties, giving conservationists a comprehensive view of the
region's ecological diversity.
The results indicate that aerial photography combined with GIS enables a precise
classification of landscape features, offering a robust tool for managing wildlife habitats. The
study reveals significant habitat diversity in subtropical estuarine environments, underscoring the
importance of tailored conservation strategies for each habitat type.
In the study of distribution of macrophyte species and habitats in South African estuaries
by Adams et al. (2016) presented another application of GIS in mapping macrophyte species and
habitats across South African estuaries. Their study highlighted the effectiveness of spatial tools
in identifying ecological zones and characterizing the distribution of aquatic plant species. The
research emphasized how GIS and remote sensing (RS) techniques could provide a baseline for
conservation planning by monitoring environmental changes over time. Such approaches are

12
highly relevant for biodiversity studies, particularly in aquatic and coastal ecosystems, where
rapid habitat changes often occur due to anthropogenic and climatic pressures.
2.1.2 Integrating GIS and Remote Sensing for Wildlife Management
Integrating GIS and RS technologies into wildlife ecology and conservation management
has revolutionized how biodiversity is studied, monitored, and protected. These tools offer
spatial and temporal insights essential for addressing complex ecological challenges and
formulating effective conservation strategies. Drawing on the foundational principles outlined by
Fryxell et al.(2014) , alongside the methodologies explored by Acharyya et al.(2017) and Prasad
et al. (2015), this section highlights how GIS and RS enhance biodiversity conservation efforts.
Fryxell et al. (2014), in their foundational text Wildlife Ecology, Conservation, and
Management, emphasize the need for a comprehensive understanding of ecological principles to
support conservation efforts. They highlight the importance of studying species distributions,
migration patterns, and habitat use in the context of broader ecological processes. GIS and RS
are pivotal in achieving these goals by enabling precise spatial analyses. Fryxell et al. (2014)
stress the value of integrating ecological models with geospatial data to predict population
dynamics and assess the impacts of environmental change on biodiversity. For instance, spatial
tools can help delineate critical habitats and identify potential corridors that maintain genetic
flow between fragmented populations. Furthermore, the authors underline the necessity of

13
addressing human-wildlife conflicts, advocating for conservation strategies informed by detailed
spatial data, such as land-use maps and human activity patterns.
Acharyya et al. (2017) explore the practical applications of GIS and RS in wildlife
management and showcase their effectiveness in monitoring and protecting biodiversity. Their
research demonstrates how satellite imagery and spatial analysis can detect deforestation, track
habitat loss, and identify areas vulnerable to poaching. For example, they highlight a case study
where satellite data was used to monitor illegal encroachments in a wildlife reserve, enabling
timely interventions to mitigate habitat destruction. Acharyya et al. (2017) also emphasize the
utility of GIS in mapping species distributions and analyzing habitat suitability, which is critical
for reintroducing species into restored habitats. Their findings underscore the role of geospatial
technologies in enabling proactive conservation measures by identifying priority zones for
intervention and resource allocation.
Prasad et al. (2015) focus on integrating GIS and RS for biodiversity conservation,
particularly in habitat monitoring and species distribution modeling. They discuss advanced
remote sensing techniques, such as high-resolution satellite imagery and vegetation indices like
the Normalized Difference Vegetation Index (NDVI), used to assess habitat quality and detect
changes over time. Their research highlights how GIS-based spatial modeling has been applied
to delineate biodiversity hotspots and evaluate the connectivity of wildlife corridors. For
instance, they document a case study in which habitat fragmentation was analyzed using
landscape metrics derived from RS data, providing actionable insights for corridor design. Prasad

14
et al. (2015) also emphasizes the role of temporal analysis in capturing seasonal variations in
species distribution, which is crucial for developing adaptive conservation strategies.
These frameworks and technologies combined offer a comprehensive approach to
wildlife conservation, balancing ecological theory with practical management strategies. This
research seeks to provide actionable insights that support long-term biodiversity preservation and
address pressing conservation challenges by leveraging spatial data and remote sensing.
2.1.3 GIS in Wildlife Tracking
GIS technologies have significantly advanced wildlife monitoring and conservation
efforts. Researchers can gain unprecedented insights into animal behavior, habitat use, and
migration patterns by combining spatial data with modern tracking systems.
Jin et al. (2023) present an integrated animal tracking system that combines GPS
technology with Unmanned Aerial Vehicles (UAVs). This system enables real-time tracking of
animal movements, offering critical insights into their spatial ecology. UAVs complement GPS
tracking by providing a bird's-eye view of an animal's surrounding environment, which is
particularly useful in monitoring species in remote or inaccessible areas. By integrating these
data into GIS, researchers can analyze movement patterns, identify habitat preferences, and
detect environmental features influencing animal behavior. Jin et al. (2023) emphasize the
efficiency of this system in reducing tracking costs while improving data accuracy, making it a
valuable tool for large-scale wildlife studies.

15
In another case by Ramesh et al. (2021), they explored a cost-effective animal location
tracking system for farm animals, utilizing Arduino and GPS modules. Arduino, an open-source
microcontroller platform, enables the creation of low-cost, customizable tracking devices by
integrating sensors and communication modules. Although developed for agricultural purposes,
this system demonstrates the scalability and adaptability of GPS-based tracking for broader
applications, including wildlife monitoring. Their study highlights how such technologies can
monitor animal locations, mitigate human-wildlife conflict, and ensure efficient resource
management. When integrated with GIS, the data collected by these systems can provide detailed
spatial insights, aiding in developing habitat preservation and conflict resolution strategies.
2.2Web Map Design for Animal Conservation
Web maps play a pivotal role in animal conservation by providing dynamic platforms for
visualizing and analyzing critical wildlife data. These digital maps integrate GIS with interactive
web technologies, enabling conservationists to monitor wildlife habitats, track migration
patterns, and detect ecosystem changes. By layering data such as vegetation cover, migration
routes, and human activities, web maps offer comprehensive insights for data-driven decisionmaking in habitat preservation and wildlife management.
2.2.1 A New Way to Map with Conservation Data
AgriLife Today (2024) explores innovative ways of using GIS and web mapping
technologies for conservation efforts in “A New Way to Map with Conservation Data”. The

16
focus is on a new method of mapping that integrates extensive conservation data into userfriendly web maps. This approach enhances conservationists' ability to visualize and manage
data concerning endangered species, habitats under threat, and areas requiring immediate
environmental protection.
The article highlights how these advanced mapping tools facilitate more informed
decision-making through better data accessibility and enable real-time collaboration across
various users. This is particularly important in conservation efforts, where timely information
can lead to more effective responses to environmental threats. The integration of interactive webbased platforms allows users to manipulate data layers and customize views to suit specific
project needs, thereby improving the efficiency of conservation planning and monitoring.
Furthermore, AgriLife Today (2024) discusses how these technologies are being adopted
by conservation agencies to engage the public, educate communities about local and global
conservation issues, and encourage more active participation in conservation initiatives. The
article underlines the transformative potential of these mapping technologies in bringing about a
more informed and collaborative approach to environmental stewardship.
2.2.2 Advanced Species Mapping for Conservation
Web-based platforms have significantly enhanced researchers' ability to visualize,
analyze, and share wildlife data. These platforms bridge the gap between technical data

17
processing and user-friendly interfaces, empowering a diverse audience of scientists,
policymakers, and the public to engage with ecological information.
LaZerte et al. (2017) introduced feedr and animalnexus.ca, a paired R package and web
application that enable the transformation and visualization of animal movement data from static
monitoring stations. The feedr package processes large datasets, automating the cleaning and
transformation of complex wildlife monitoring data. Meanwhile, the animalnexus.ca platform
allows users to interact with these data through a web-based interface, providing visualizations
that highlight movement patterns, site usage, and temporal trends. This dual approach is
particularly advantageous for managing datasets generated by long-term monitoring programs, as
it combines the computational power of R with an intuitive interface that broadens accessibility.
LaZerte et al. (2017) emphasize the importance of user-focused design in ensuring that these
tools are adequate for researchers and the public, making ecological insights actionable.
In another study by Dwyer et al. (2015) developed an open, web-based system
specifically designed to analyze and share animal tracking data. The platform integrates multiple
datasets, allowing researchers to analyze spatial and temporal patterns across various species and
geographic scales. A key feature of their system is its ability to support collaboration among
researchers and conservation practitioners, promoting transparency and encouraging multistakeholder involvement in data interpretation. The system also includes tools for filtering and
visualizing movement data, enabling users to identify patterns such as migration corridors,
hotspots of activity, and temporal shifts in habitat use. This approach aligns with modern

18
conservation practices, prioritizing data sharing and collective decision-making to address
complex ecological challenges.
2.3Volunteered Geographic Information in Wildlife Conversation
VGI is an emerging data source that has transformed how wildlife conservation projects
gather and utilize information on species distribution and movement. Platforms such as
iNaturalist enabled millions of users to contribute geospatially referenced data on species
observations, leading to massive datasets that would otherwise be unfeasible to collect through
traditional fieldwork alone. However, integrating VGI into formal research frameworks brings
about concerns about data quality, accuracy, and the uncertainty that arises from user-generated
content. This section explores various aspects of VGI in wildlife conservation, focusing on
managing uncertainty and integrating VGI with formal datasets.
2.3.1 Uncertainty-Aware Enrichment of Animal Movement Trajectories by VGI
One innovative approach to overcoming the challenges associated with VGI is presented
by Hartmann et al. (2022) which demonstrates a method to enrich GPS-based animal movement
data with VGI contributions, increasing the spatial and temporal scope of datasets used in
movement ecology. The researchers highlight the potential of combining biologging data from
GPS-collared animals with user-contributed data from platforms such as iNaturalist and eBird.
The study develops an interactive system called BirdTrace (“VGIscience | Birdtrace,” n.d.),
which facilitates the integration of VGI data with GPS-tracked bird movements while accounting

19
for the inherent uncertainty in citizen-contributed data. The key innovation of this system is the
uncertainty-aware framework that visually and statistically highlights the variability and
potential inaccuracies in VGI contributions, such as species misidentification or incorrect
geolocation. By implementing an uncertainty metric, researchers can assess the reliability of each
VGI contribution before integrating it with more precise biologging data.
This approach directly relates to this project's goals of integrating VGI from iNaturalist
with formal data collection and processing systems. The researchers in this case merged GPS
data with VGI, aimed to incorporate new wildlife observation records with historical observation
data while accounting for uncertainties in the VGI dataset. Additionally, the location-blurring
techniques used in this project to protect endangered species align with the uncertainty-aware
enrichment model, where sensitive or less specific data is handled carefully to avoid misuse. This
hybrid approach, which combines the strengths of VGI and traditional data sources, can lead to
more robust, actionable insights for wildlife conservation.
2.3.2 VGI and the Spatial-Discursive Construction of Nature
In the study by Astaburuaga et al. (2022), the authors investigate how maps and VGI not
only serve as tools for geographic representation but also function as discursive instruments that
shape human perceptions of nature. Focusing on the Patagonia-Aysén region in Chile, the study
reveals that digital mapping and VGI contribute to constructing nature as "pristine,"
"untouched," and often commodified for tourism and conservation efforts. The research

20
highlights that these tools are far from neutral; instead, they reflect and reinforce specific cultural
and economic narratives, particularly within nature-based tourism.
Astaburuaga et al. (2022) argue that VGI contributes to the creation of a certain "spatial
imaginary" of nature, often aligned with dominant discourses that frame natural spaces as
commodities for human consumption, either for recreation or conservation. This understanding
of nature can influence how society engages with and manages natural resources, including
wildlife. The study emphasizes the importance of critically analyzing the power of spatial media,
as these tools can perpetuate views of nature while marginalizing others.
When VGI is integrated into conservation mapping, as with wildlife monitoring systems
like iNaturalist, it plays a dual role: providing valuable data for tracking species distributions
while contributing to societal narratives about nature. Therefore, the implications of VGI for
both scientific analysis and public engagement should be carefully considered to ensure that
these maps foster a more nuanced understanding of biodiversity and ecological processes.
Furthermore, Astaburuaga et al. (2022) focused on the potential biases and uncertainties
inherent in VGI data resonates with ongoing efforts to refine data collection methodologies for
wildlife conservation. Enabling location-blurring techniques to protect endangered species helps
address the ethical concerns of making sensitive spatial data publicly available.

21
2.3.3 Assuring the Quality of Volunteered Geographic Information
Goodchild and Li (2012) provide a foundational framework for assessing VGI quality,
identifying key dimensions such as positional accuracy, attribute accuracy, completeness,
consistency, and lineage. Their study highlights the unique challenges associated with ensuring
data quality in crowdsourced geographic information due to the decentralized and voluntary
nature of VGI contributions. Goodchild and Li (2012) propose a combination of intrinsic and
extrinsic evaluation methods to mitigate quality issues. Intrinsic methods include analyzing the
consistency and patterns within the VGI dataset itself. In contrast, extrinsic methods involve
cross-referencing VGI with authoritative data sources, such as government or scientifically
validated datasets.
A study by Flanagin and Metzger (2008) delves into the role of user reputation and
validation mechanisms in enhancing VGI reliability. The authors argue that building user
profiles based on expertise, experience, and accuracy history can help assess the reliability of
individual contributions. For instance, experienced users with a track record of accurate
submissions could be given higher credibility scores, and their data can be weighted more
heavily in analyses. Many VGI platforms, including iNaturalist, have adopted peer review
systems that allow other users to validate or question specific observations, thereby enhancing
data reliability.

22
Other methods to assure VGI quality involve automated validation techniques. Haklay
(2010) suggests using algorithms to detect outliers and unusual patterns in VGI submissions,
which may indicate potential errors or inaccuracies. By employing machine learning models to
flag data that diverges significantly from known distributions, automated systems can assist in
maintaining a consistent level of quality across large datasets.
Despite these quality assurance efforts, the inherent uncertainty in VGI remains a
challenge for conservation applications. Strategies such as cross-verifying VGI with remote
sensing data, restricting observation updates to reduce potential threats (such as poaching), and
employing peer validation methods aim to balance data accessibility with reliability. These
measures not only enhance the quality of VGI but also address ethical considerations by
minimizing the potential misuse of sensitive wildlife data.

23
Chapter 3: Methods
This chapter outlines the methodological framework employed in the development of this
project. The methods described in this section are centered on data collection, processing, and
integration within a GIS environment, as well as the design and deployment of the web
application. Key processes include the gathering of wildlife data from iNaturalist, cleaning and
refining the data to ensure accuracy and privacy, and visualizing the information using ArcGIS
tools. Additionally, a survey for new wildlife observations was designed to facilitate user
interaction. Each step of the methodology is aimed at producing an informative, user-friendly
platform that supports wildlife conservation efforts.
3.1 Data Collection and Processing
The foundation of the "Historical Observation of Volunteered Geographic Information
Data Web Map for Wildlife in Kenya" project is built upon robust data collection and processing
methodologies, designed to ensure the accuracy, relevance, and usability of the spatial data used
in this study. This section details the processes involved in sourcing, cleaning, and preparing the
data for integration into a GIS web application.
3.1.1 Data Source: iNaturalist
iNaturalist is the key data source in the project, it is a public science platform that allows
individuals to record, share, and discuss biodiversity observations. Founded in 2008, iNaturalist

24
has grown into one of the largest databases for biodiversity data, with millions of contributions
from users worldwide (iNaturalist.org 2025). The platform’s main appeal lies in its ability to
engage a wide community of users—ranging from professional scientists to hobbyists—in
documenting species across the globe. Its open-source nature and ease of use have made it a
valuable tool for both educational purposes and scientific research.
iNaturalist functions by allowing users to upload observations of plants and animals,
which can include photographs, geographic coordinates, and temporal data. These observations
are then either identified by the community or using artificial intelligence models. Once multiple
users verify the observation, it can reach "research-grade" status, which means it can be used for
scientific and conservation purposes. Many of these verified observations are then shared with
global biodiversity databases, such as the Global Biodiversity Information Facility (GBIF),
further extending their utility.
iNaturalist implements specific measures to handle endangered species data responsibly.
Precise location data is automatically obscured for species flagged as threatened or endangered
based on IUCN Red List (2024) criteria or regional conservation designations. Instead of sharing
exact coordinates, iNaturalist replaces them with randomized points within a larger area to
protect sensitive habitats from potential exploitation. Trusted researchers or organizations can
access precise data after demonstrating a legitimate conservation need and agreeing to safeguard
its use. These practices align with this project’s approach to balancing data accessibility and
species protection.

25
Understanding iNaturalist's capabilities and limitations is crucial to integrating it
effectively as a primary data source for this project. To evaluate iNaturalist, a strengths,
Weaknesses, opportunities, and Threats (SWOT) analysis (Table 1) was conducted. The SWOT
analysis provides a comprehensive framework for evaluating iNaturalist's role in this project. By
identifying its strengths, such as its large user base and multi-platform compatibility, the table
highlights the aspects that make iNaturalist a valuable data source. At the same time, it considers
weaknesses like limited filtering options and potential user confusion due to simplistic
symbology. The analysis also identifies opportunities for growth, such as collaborations with
conservation organizations, while acknowledging external threats, such as the challenge of
securing consistent funding.

26
Table 1. SWOT Analysis for iNaturalist
Strength Weakness
• Large User Base: iNaturalist has a
large community of users contributing
observations and data.
• Compatible with Multiple Platform:
iNaturalist can be used on PC, tablets
or phone. Data can be used across
different platforms.
• Data Orientation and visualization:
iNaturalists does not support user to
filter specific species on their web
map, which leads to long loading time
and distractions when visualizing data.
• Simple Symbology: iNaturalist uses
the simplest location icon to represent
all data points and differ them with
colors. But this leads to confusion
since some colors are very similar.
Opportunity Threats
• Collaboration with Institutions:
Partnering with academic institutions
and conservation organizations can
enhance data validation and research
collaboration.
• Public Awareness: Growing interest
in biodiversity conservation globally
offers opportunities to raise public
awareness through the platform.
• Funding and Sustainability:
Ensuring consistent funding to
maintain and develop the platform is
crucial and can be a significant
challenge.

27
Since the SWOT analysis aims to leverage strengths and opportunities while addressing
weaknesses and threats, Table 2 outlines potential solutions to the weaknesses and threats
identified in the previous analysis. These solutions aim to ensure iNaturalist’s effectiveness and
reliability as a tool for wildlife conservation while addressing its limitations in data visualization
and sustainability.
Table 2. Potential Solutions to iNaturalist’s Weakness and Threat
Solutions to Weakness Solutions for Threat
• Data Orientation and visualization:
This web application only focus on
the four chosen taxa and allow users
to turn on and off irrelevant layers as
needed.
• Simple Symbology: Cartoon style
and more diverse colors are employed
to differ different taxon group.
Clustering allows users to observe
where wildlife gathering directly.
• Funding and Sustainability: This
project is developed completed based
on personal interest and will never aim
for any profitable target. This project
is not aim for any profitable goals all
time.
3.1.2 Data Extraction
Data extraction from iNaturalist was conducted using the platform's Application
Programming Interface (API), which allows for efficient querying and retrieval of large datasets.
Specific filters were applied to ensure that only relevant data was collected. These filters
included:
• Taxon Identification: Only data classified under the four selected taxa (Amphibians,
Mammals, Aves, and Reptiles) were extracted.

28
• Geographical Boundaries: Observations were limited to those within the political
boundaries of Kenya.
• Temporal Range: The data was restricted to the period between January 1, 2010, and
December 31, 2020.
• Data Quality: Only observations classified as "research grade" were included, ensuring a
higher level of accuracy. Research grade observations on iNaturalist are those that have
been confirmed by two or more users and include accurate geolocation data.
Table 3 provides a comprehensive overview of this project's data types, sources, and
specific purposes. This table highlights the observation data extracted from iNaturalist and the
supplementary vector data collected from credible sources to enhance the spatial analysis. These
datasets form the foundation for analyzing wildlife patterns, understanding spatial distributions,
and contextualizing the ecological and anthropogenic factors influencing the study area.

29
Table 3. Data Types and Purposes in the Project
Data (Count of Data) Source Description and Purpose
Amphibia Points (492) iNaturalist Geocoded data with animal
locations, time observed, images
etc. To present historical
locations of animals, and to
analyze animal pattern through
ArcGIS tool such as Hotspot
Analysis.
Aves Points (24215)
Mammalia Points (17278)
Reptilia Points(2912)
Kenya Country Border Africa Geoportal Boundary of the study area
Kenya Wetlands and
Protected Areas
National GeospatialIntelligence Agency
Polygons of Kenya wetlands and
protected areas
Kenya Roads and City
Boundaries
OpenStreetMap Outlines of main traffic roads and
city boundaries.

30
3.2 Geographic Information System (GIS) Integration
ArcGIS Pro is employed in this project as the primary tool for visualizing wildlife
observation data collected across Kenya. By leveraging the platform’s advanced spatial analysis
and mapping capabilities, the project transforms raw data from sources such as iNaturalist into
interactive, informative maps that highlight species distributions and biodiversity patterns.
The data visualization process in ArcGIS Pro begins with the systematic import, cleaning,
and management of spatial datasets. Once imported, the data is organized into customized layers
representing different taxonomic groups and geographical areas. Using the platform’s robust
symbology and analysis tools, these layers are styled to create clear, user-friendly visual
representations of wildlife distributions. The final step involves preparing the data for integration
into web-based applications, allowing for interactive exploration by researchers, policymakers,
and the public. This section details the procedures undertaken in ArcGIS Pro, from data
management to the creation of dynamic, web-ready maps that support conservation efforts.
3.2.1 Data Import
The integration of GIS data begins with importing the raw datasets from various sources,
including iNaturalist VGI data and map elements such as Kenya boundaries. Effective data
management is essential to ensure that the imported data is correctly formatted, stored, and
organized for subsequent analysis and visualization. This section outlines the procedures

31
followed for importing data into the GIS environment, managing the database, and organizing
data layers to maintain consistency and usability.
The wildlife observation data used in this project was sourced from iNaturalist, provided
in s CSV. ArcGIS Pro allows for seamless import of these formats, converting them into GISreadable data structures that can be further processed and visualized.
• CSV Import: CSV files containing wildlife observation data were imported into ArcGIS
Pro using the XY Table to Point function as shown in Figure 2. This feature converts
tabular data with latitude and longitude values into point feature classes, enabling
geographic visualization. Each record represents a unique wildlife observation, with
attributes such as species name, timestamp, and observer details.
Figure 2. XY Table to Points Tool

32
• Vector Data (Basemap) Import: The country boundary of Kenya was imported from the
Esri Africa Geoportal (2023). The roads and city locations were extracted from
OpenStreetMap. After comparing and checking the background information between
different datasets, most recently updated credential datasets were imported to enrich the
basemap.
3.2.2 Data Cleaning
Upon importing, the data underwent a rigorous cleaning and pre-processing phase to
prepare it for analysis and integration into the GIS. This process involved several steps:
• Duplication or Blank Data Removal: Entries were flagged as duplicates if they shared
identical taxon identification, timestamp, and geographic coordinates. These records were
then systematically compared and merged or removed to maintain a unique set of
observations for each observation. For instance, duplicate reports of the same animal in
the same location and time window were consolidated into a single record. Entries
missing key attributes such as geolocation, taxon classification, or timestamp were
considered incomplete and were removed from the dataset. In cases where taxonomic
data or timestamps were present but lacked geographic coordinates, additional validation
was attempted via referencing other records, but if unresolved, these entries were
discarded.

33
•Geolocation Accuracy: Each geographic entry was checked for plausible values in
terms of latitude and longitude, ensuring that all points fell within the expected
boundaries of the study area (Kenya). Points located outside the region or placed
obviously incorrectly (e.g. Crocodile in the desert) were removed. All the data are
reprojected into WGS 84.
3.2.3 Hotspot Analysis
Hotspot analysis was employed as a key method for identifying areas with high
concentrations of wildlife observations across the study region. This technique utilizes GIS tools
to evaluate spatial clustering patterns within the historical dataset (2010–2020), highlighting
critical regions for wildlife conservation. The Getis-Ord Gi* statistic was applied to detect
statistically significant clusters of high and low wildlife densities, revealing spatial trends that
inform conservation strategies. These hotspots provide users data-driven insights into
biodiversity trends (Garrah 2015).
3.2.4 Data Management
Once the data was successfully imported and cleaned, it was stored and managed within a
geodatabase to streamline data handling and enhance performance. The geodatabase acts as a
centralized repository for the project’s spatial data, allowing for efficient querying, updating, and
managing of datasets.

34
• Layered Structure: Separate layers were created from the raw wildlife dataset for each
taxonomic group (Amphibians, Mammals, Aves, and Reptiles), ensuring that the data
could be visualized and analyzed individually or in combination. Each layer contains the
spatial locations of the respective taxon’s observations along with the associated
attributes (e.g., species name, observation date). Meanwhile, other map contents such as
basemap content and hotspot analysis were grouped individually for better organization
• Geodatabase: The use of a file geodatabase was chosen for this project due to its ability
to store large datasets, maintain relationships between spatial and non-spatial data, and
provide the necessary scalability for future updates. The geodatabase contains feature
classes for spatial data and tables for non-spatial attributes, such as observer details or
metadata on each observation.
• Local Backup: To keep the web application functional and avoid losing connection to
online data, a backup database and ArcGIS Pro Package were created on a personal
device and ready to transfer to another account in the future if needed.
3.3 Experience Builder
As part of the data visualization process, ArcGIS Experience Builder, Enterprise Version
(2024), was employed to create an interactive web-based platform that allows users to explore
wildlife observation data. ArcGIS Experience Builder is a flexible tool that enables the
development of custom web applications with dynamic maps, rich user interfaces, and interactive

35
features. This section outlines the steps including data exporting, web map design and
implementation using ArcGIS Experience Builder, focusing on user interaction, layout design,
and functionality.
3.3.1 Data Export to ArcGIS Online
Before exporting data from ArcGIS Pro to ArcGIS Online, data preparation within
ArcGIS Pro involves ensuring that all layers are correctly structured, symbolized, and contain
accurate geospatial references. This process is described within the previous sections.
Following data preparation, the next step is publishing the data as a web layer. In ArcGIS
Pro, users can right-click the target layer, browse the "Share" tab, and choose "Publish Web
Layer." This process involves configuring several settings, such as naming the web layer, adding
descriptions, and defining layer types. Users also define who can access the layer through
sharing settings, choosing between public access, restricted organizational access, or groupspecific access.
Once the web layer is published to ArcGIS Online, the next step is to verify and group
the data. This verification process ensures that all data attributes, symbology, and pop-ups have
been correctly transferred and appear as intended. By navigating to the "Content" tab in ArcGIS
Online, users can review and interact with the published web layer to ensure that it behaves as
expected in the online environment. In the end, a folder called "594 Thesis Project" is created as
shown in Figure 3, which holds all the data employed in this project.

36
Figure 3 ArcGIS Online Layer Management
3.3.2 Web Map Design
This section focuses on refining the map content after data is exported from ArcGIS Pro
to ArcGIS Online. The first step is the selection and customization of a basemap. The basemap
provides the geographical context for the wildlife data. A simple, clean basemap is essential to
avoid overwhelming users with excessive information. In this project, a light gray canvas
basemap was chosen for its minimalistic design as shown in Figure 4, allowing wildlife layers to
stand out.

37
Figure 4. Basemap Editing
Once the basemap design is finished, the next step is to add the data layers exported for
ArcGIS Pro to the map. By clicking “Add” in the layer tab, at the bottom of Figure 5, and users
can choose the source of data. After navigating to “My Content,” all the related data are added.

38
Figure 5. Adding Map Contents
With many data points, particularly for taxa and many observations, the clustering
function in Figure 6 can significantly improve the visual effect. Clustering groups observations
geographically close into a single cluster, with a number indicating the count of observations in
that area. This reduces visual clutter and improves the map's readability, especially when zoomed
out. As users zoom in, the clusters break apart into individual points, allowing for detailed
exploration of specific observations.

39
Figure 6. Data Clustering Setting
In the meantime, the original pop-up windows contain too much irrelevant information,
such as the name of the users who created the observation. Therefore, as shown in Figure 7, popup windows are configured only to show key attributes like species information, observation
date, and an image (if available). These pop-ups provide an interactive way for users to access
additional details about each observation. The pop-up window’s design ensures the information
is well-structured, visually appealing, and easy to navigate. Figure 8 provides an example of popup window of cheetah.

40
Figure 7. Pop-ups Setting

41
Figure 8. Example of a Pop-up Window

42
A time slider is added to the map (Figure 9). This tool allows users to filter observations
by date, providing insights into temporal patterns in wildlife observations across the 2010-2020
dataset. Users can explore how wildlife distributions have changed over time, which can be
valuable and efficient for researchers and conservationists. The default temporal range is set to
be three months, a season-long for effective observation, starting from January 1st, 2010.
Figure 9. Time Slider Setting

43
3.3.3 Integration in Experience Builder
The ArcGIS Experience Builder is one of the most important elements in this project. It
not only holds the wildlife web map but also provides multiple widgets that enhance public
interactivity. The layout of the Experience Builder is designed to make the map the central focus
of the application, ensuring users can easily interact with the wildlife data.
Different layout sizes were employed to ensure that the application adapts seamlessly to
different screen sizes, from desktop (Figure 10) to mobile devices (Figure 11).
Figure 10. Experience Builder Map Appearance

44
Figure 11. Map Content Appearance on Small Screen Device

45
A search box widget is integrated into the Experience Builder to enable users to locate
specific regions or species within the map. This widget supports filtering by taxonomic groups or
periods, allowing users to tailor their exploration and focus on relevant data. Figure 12 illustrates
the back-end configuration view of the search box widget. This view displays the setup options
available to the developer, such as selecting data sources, enabling filtering for specific taxa
(e.g., Amphibia, Aves, Mammal, Reptile), and configuring search fields like genus, common
name, scientific name, and species guess
Figure 12. Search Box Design

46
3.4 New Observation Report
The new observation report feature allows users to contribute wildlife documentation
through a structured data submission form built using Esri Survey123 (Figure 13). This tool
enables the collection of valuable wildlife data, including species identification, geographic
location, observation time, and optional photographic evidence. However, to prevent the misuse
of sensitive wildlife location data, such as for poaching, these observations are not integrated into
the web map in real time. Instead, all submitted observations are securely stored and reviewed
periodically to ensure data accuracy and conservation integrity. Unlike conventional citizen
science platforms, where data is immediately accessible, this project follows a scheduled update
cycle to protect endangered species and minimize ecological risks. New observations will be
incorporated into the public dataset every five years, aligning with the project's commitment to
responsible data sharing.

47
Figure 13. Esri Survey123 Form Appearance
The structure of the Survey123 form is divided into several key fields, each carefully
designed to gather critical information related to the wildlife observations. These fields are
required or optional based on the nature of the data needed, with clear prompts to guide users
through the process. Figure 14 shows the creation panel of the Esri Survey123.

48
Figure 14. Esri Survey123 Creation
The survey includes the following questions:
• Location of Observation: The survey form includes a map-based geolocation feature to
capture precise geographic data for each observation. By default, the form automatically
records the user's location using their device's GPS. However, users can manually adjust
their location by dragging the point on an interactive map. This flexibility ensures that
observations are accurately geolocated, even in areas where GPS signals may be weak or
imprecise. Allowing users to refine their location improves data quality, especially in
regions with low locational accuracy or when observations are made near spatial
boundaries. This is a required question.

49
• Taxon Observed: This field provides users with a dropdown list of taxa, categorized into
the relevant taxonomic groups (Amphibians, Mammals, Aves, and Reptiles). This is a
required question.
• Genus Or Species Observed: This field encourages users to identify the specific species
they reported. However, users can also guess if they are unable to identify the specific
creature. This is an optional question.
• Date and Time of Observation: Users are required to input the date and time of their
wildlife observation. This information helps in tracking temporal patterns in wildlife
activity and understanding seasonal or diurnal behavior. This is a required question.
• Notes: An optional field allows users to provide additional context to their observation,
such as the behavior of the animal, the environment when they observed the animal, or
any notable conditions (e.g., weather). This qualitative data can offer further insights into
the wildlife behaviors being observed.
• Media Attachments: Users are encouraged to upload media files such as photos or
videos of the observed species. This visual evidence aids in the verification process,
especially for observations of rare or endangered species.
• Contact Information: Optional fields regarding phone number and email are provided if
users expect further conversation.

50
3.5 Web Framework
The Historical Kenya Wildlife Observation Map web application serves as an interactive
platform for observation, reporting, and visualizing wildlife observations in Kenya. This section
outlines the design and development of the web frame, focusing on the integration of wildlife
data with a user-friendly, intuitive interface. The primary objective of the web frame is to
facilitate public participation while ensuring that data is responsibly accessed and used.
The web frame was designed using a combination of HTML and CSS and divided into
several sections, including a header with additional resources links, an iframe container for
viewing the ArcGIS-based wildlife map, a link for submitting new wildlife reports, and a footer
with disclaimers and attribution.
3.5.1 Welcome Message Pop-up
The pop-up window of the welcome message is the first interactive function that provides
users with an introductory message and a legal disclaimer. This message in Figure 15 ensures
that users understand the purpose of the site, the terms of use, and the importance of responsible
behavior, particularly given the sensitive nature of the wildlife data.

51
Figure 15. Welcome Message Pop-up
The welcome message pop-up is implemented using a modal window (Figure 16), and it
is triggered automatically when the page loads and displays a message that welcomes the user to
the web application, followed by a brief introduction and legal disclaimer. Users must click the
"Agree" button to close the pop-up and acknowledge the terms, thus ensuring they understand
the usage restrictions before interacting with the site.

52
Figure 16. HTML Expression for the Pop-Up
When the page finishes loading, the openModal() function is called, which sets the
modal’s display property to flex, making it visible to the user. The modal remains visible until
the user clicks the "Agree" button. When the user clicks the "Agree" button, the closeModal()
function is triggered, setting the modal’s display back to none, effectively hiding the modal and
allowing the user to proceed to the main content (Figure 17).
Figure 17. JavaScript Expression for the Pop-up Modal

53
In Figure 18, the modal is centered using Flexbox properties (justify-content: center;
align-items: center), which vertically and horizontally centers the content within the viewport.
This layout ensures that the modal remains central and responsive across all screen sizes.
Figure 18. CSS Expression for the Welcome Message

54
The "Agree" button is styled to align with the overall theme of the web application, using
the same green color scheme (background-color: #4CAF50) as the header and other buttons. The
button is centrally aligned and provides a clear call to action for users to acknowledge the
disclaimer before they proceed to use the application. A hover effect (background-color:
#45a049) is added to notice the users they have selected the button.
3.5.2 Header, Additional Resources Bar and Help Icon
The header (Figure 19) is a critical part of the web page as it establishes the identity of
the application and provides navigational elements to guide users through the site. It plays a
central role in maintaining a professional look and ensures that key information is easily
accessible to the user from the moment they land on the page.
Figure 19. Header Appearance
• Header
The header is designed using the HTML element (Figure 20), which
semantically defines the introductory section of the web page. It houses the application title,
additional resources links, and a help icon (question mark button). The layout of the header is

55
controlled through CSS Flexbox to ensure that the elements align and respond to different screen
sizes while maintaining a clean and structured appearance.
Figure 20. HTML Expression for the Header and Part of Elements in the Container
The code in figure 21 shows the styling setting of the header. The title "Historical Kenya
Wildlife Observation Map" is centered using the Flexbox justify-content: center property, which
places the title evenly across the available space. This ensures that regardless of screen size, the
title is always placed at the center of the header.

56
Figure 21. CSS Expression for the Header
The title is styled with a font size of 1.5rem and bold text (font-weight: bold), making it
stand out against the background while remaining accessible and readable across devices.
The header uses a green background (#4CAF50) to create a visually pleasing look that
aligns with environmental and wildlife themes. This also helps distinguish the header from the
rest of the page content. The text within the header is white (color: white), providing sharp
contrast and enhancing readability.

57
• Additional Resources
The additional resources links are located on the right side just below the title in the
header, contains links to external resources and relevant information. These links allow users to
quickly access wildlife-related websites, such as Wildlife in Kenya, Animal Protection, and AntiPoaching.
Additional Resources Link Styling: Each link is styled as a block-level element with
padding (padding: 8px 12px) and a background color of dark gray (#333). The links change color
when hovered over (background-color: #555), providing visual feedback to users and
encouraging interaction (Figure 22). The hover effect is achieved using the CSS :hover pseudoclass, which smoothly transitions the background color, enhancing the overall user experience.
Figure 22. CSS Expression for Additional Resources Bar

58
The additional resources bar is responsive, adjusting its layout depending on the screen
size. For larger screens, the links are displayed side-by-side in a horizontal row. On smaller
screens, the Flexbox flex-wrap property ensures the links stack vertically or wrap, preventing
overlap and maintaining usability on mobile devices.
• Help Icon (Question Mark)
The help icon is positioned using absolute positioning as shown in Figure 23 (position:
absolute; right: 20px; top: 50%;) to ensure it remains at the top-right corner of the header,
regardless of the screen size. The transform: translateY(-50%) property is used to vertically
center the icon relative to the height of the header.
Figure 23. CSS Expression for the Help Icon

59
When hovered over, the button displays a tooltip with the message "How it works?". This
is implemented using the CSS ::after pseudo-element, which generates the tooltip only when the
user hovers over the icon, making the design interactive and user-friendly.
The tooltip is styled with a light gray background (background-color: #555) and white
text, providing a subtle yet noticeable guide for the user without cluttering the interface.
3.5.3 Container
The container of the web page in Figure 24 is designed to hold the primary interactive
elements, such as the ArcGIS map iframe and the "New Observation" button. This section plays
a pivotal role in presenting data to users in a structured and responsive manner, ensuring that the
wildlife observation data is easily accessible and visually appealing. The layout of the container
is handled with CSS Flexbox, ensuring the section adapts smoothly across different device
screen sizes.
Figure 24. Maps and Other Elements in the Container

60
The container section is implemented using the HTML element with the class
container. It is a flexible container that adapts the layout depending on the content inside it,
making it well-suited for housing the interactive elements of the site (Figure 25).
Figure 25. HTML for the Container and “New Observation” Button
The Flexbox layout model is used to ensure the internal content is arranged in a manner
that adapts to varying screen widths. The flex-direction: column property ensures that all the
elements (the additional resources links, iframe, and "New Observation" button) are stacked
vertically in a column. The use of Flexbox allows the layout to adjust dynamically based on the
available screen space, providing a seamless user experience regardless of device (Figure 26).

61
Figure 26. CSS Expression for the Container Structure and Layout
The container is given padding (padding: 10px), which helps maintain spacing between
the contents and the edges of the screen. This creates a clean and visually appealing layout,
enhancing the readability and usability of the page elements.

62
The ArcGIS map iframe is embedded within a separate container to ensure that the map
fits within the page while maintaining its aspect ratio. This section uses an advanced CSS trick
involving padding to make the iframe responsive while preserving the integrity of the content
displayed.
In Figure 27, the iframe container uses percentage-based padding (padding-top: 50%) to
ensure that the iframe's height always remains 50% of its width, thereby maintaining a 2:1 aspect
ratio. This technique is particularly effective for making the iframe responsive, allowing it to
scale properly on both large desktop monitors and smaller mobile devices without distortion.
Also, it is positioned absolutely within the container (position: absolute), meaning it fills the
available space completely without interfering with other elements on the page. This allows the
map to occupy the entire container while adjusting to different screen sizes without introducing
scrollbars or cutting off any part of the content. Rounded corners (border-radius: 5px) are used to
soften the appearance of the iframe container, making it visually consistent with the rest of the
design elements on the page, such as the buttons and links. This helps create a cohesive and
professional look.

63
Figure 27. CSS Expression for iframe
The "New Observation" button is placed below the iframe and provides users with a way
to submit new wildlife observation records. This button links to an external survey tool, allowing
users to contribute data in a structured format.
The button is centrally aligned below the iframe, ensuring that it is both easily accessible
and visually distinct from the map. It is styled with a green background (background-color:
#4CAF50), which is consistent with the site's overall environmental and wildlife theme. The
button changes color on hover (background-color: #45a049) and slightly increases in size
(transform: scale(1.05)), providing visual feedback and encouraging interaction (Figure 28).

64
Figure 28. CSS Expression for the New Observation Link
3.5.4 Footer
Positioned at the bottom of the page, the footer is designed to remain fixed, ensuring that
the information it contains is always visible, regardless of how the user navigates the page
(Figure 29).
Figure 29. Footer Appearance
The footer section of the web application serves both a functional and informational role,
providing users with a concise disclaimer about the data used, while also contributing to the
overall aesthetic consistency of the page (Figure 30).

65
Figure 30. HTML Expression for the footer
In Figure 31, the footer is styled with a green background color (#4CAF50), consistent
with the rest of the page’s environmental and wildlife-themed design. This creates visual
continuity across the entire application and reinforces the theme of wildlife conservation. The
text within the footer is set in white (color: white), ensuring high contrast and readability. The
information conveyed includes a reference to the data source, iNaturalist, and a notice about the
frequency of database updates, which is designed to minimize poaching risks by delaying the
public release of new data.
Figure 31. CSS Expression for the Footer

66
3.5.5 Instruction Video Pop-up
The "How It Works" video pop-up in Figure 32 is designed to give users a visual tutorial
on how to navigate and interact with the web application. By embedding an instructional video,
Figure 33, directly into the web interface, users can learn the app’s functionalities, such as
submitting new observations, exploring the wildlife map, and understanding the significance of
the data displayed. The pop-up is designed to be unobtrusive and can be triggered on demand
when users click the question mark icon in the header.
Figure 32. Hovering Text for Help Icon
The video is displayed in a modal pop-up that overlays the page, ensuring that users can
access the information without being redirected to an external site. This design choice keeps
users engaged and focused on the application while enhancing their experience.
Figure 33. Instruction Video Appearance

67
This Pop-up is designed to provide users with an intuitive guide on how to navigate the
web application. The HTML structure in Figure 34 utilizes a modal component that overlays the
main content and temporarily captures the user's focus. The modal is triggered when the user
clicks on the question mark icon, located on the right side of the header, providing an immediate
way for users to access instructions.
Figure 34. HTML Expression for the Instruction Video
The modal itself is defined using a element with a unique ID (howItWorksModal),
allowing it to be targeted by JavaScript functions for both opening and closing. Inside the modal,
there is a heading that briefly describes the purpose of the video, followed by the main
content, which is an embedded YouTube instructional video. The video is displayed through an
element, which is styled to maintain a 16:9 aspect ratio across various screen sizes.
This is achieved using CSS properties such as position: relative and padding-bottom: 56.25%,
ensuring that the video remains responsive and adjusts to different devices.
Additionally, the modal includes a "Close" button at the bottom, which allows users to
exit the pop-up once they have finished viewing the video. This button triggers a JavaScript
function that changes the display property of the modal, making it invisible and returning the
user to the main application.

68
The CSS used for the "How It Works" video pop-up shares similarities with the design of
the Welcome Message pop-up (Figure 35). The .modal class is responsible for the full-screen
overlay effect, and the Flexbox (justify-content: center; align-items: center;) to center the content
both horizontally and vertically, ensuring a seamless user experience across different screen
sizes.
Figure 35. CSS Expression for the Instruction Video

69
While the structure remains consistent with the Welcome Message, the How It Works
pop-up has additional CSS for video embedding. The embedded video is styled within an iframe
using a 16:9 aspect ratio trick (padding-bottom: 56.25%), ensuring that the video is responsive
and scales appropriately across devices. This aspect is specific to the video pop-up, as it requires
careful handling to ensure the iframe maintains its correct dimensions without distorting the
layout.
This pop-up also includes a dedicated close button (.close-btn), which shares design
features with the button used in the Welcome Message. The button is styled consistently,
maintaining the app’s color scheme, and providing a hover effect (background-color: #45a049)
for a cohesive user experience.
The web framework of the Historical Kenya Wildlife Observation Map (Figure 36) was
designed to create an engaging and user-friendly platform that effectively integrates wildlife data
with public interaction. By leveraging responsive design principles and intuitive layouts, the
framework ensures accessibility across various devices and user demographics. Integrating
visual and interactive elements, such as the ArcGIS map, submission forms, and instructional
pop-ups, demonstrates a thoughtful approach to combining technological functionality with
aesthetic appeal. This framework supports the project’s broader goal of promoting public
engagement and enhancing awareness of wildlife conservation.

70
Figure 36. Historical Kenya Wildlife Observation App

71
Chapter 4. Results
This chapter discusses the results of the development and functionality of the Historical Kenya
Wildlife Observation application. Section 4.1 focuses on the interactive map, the most crucial
content for visualizing wildlife observations and exploring spatial data. Section 4.2 introduces
the web design components, including user interface elements and the application's
responsiveness across various devices. Section 4.3 covers the data collection survey integration
through Esri Survey123, explaining how new wildlife observations are reported and managed.
Each section highlights key features, design decisions, and user engagement strategies
implemented in the application.
4.1 Interactive Map
The interactive map is the core feature of the Historical Kenya Wildlife Observation
application, providing a detailed, user-friendly platform for visualizing wildlife observations
across Kenya. The map is organized into multiple layers, each representing a taxonomic group.
On the bottom of the map is a time filter (Figure 37), allowing users to focus on specific
periods within the 2010-2020 range. This temporal filter helps users to analyze changes in
wildlife distribution over time, although only historical data is shown to protect sensitive
information and prevent poaching.

72
Figure 37. Web Map Appearance
In this map, each species group is represented by distinct icons that reflect the group’s
characteristics, with varying colors to ensure ease of interpretation, as shown in Figure 38. The
map also incorporates clustering techniques to manage the display of dense data points in regions
with numerous wildlife observations. As users zoom in, clusters break apart into individual
points, allowing for detailed exploration without cluttering the display at broader scales.

73
Figure 38. A Closer look for Symbology of Data Points
Several map elements enhanced the interactive experiences. On the top right of the map,
a layer button is added so the users can turn on or off the layers they target for, which help them
to concentrate on the specific data (Figure 39). Also, a dynamic legend is included to help users
understand the symbology used for the different taxonomic layers. The legend updates based on
the layers currently active, ensuring that users can easily interpret the data they are viewing.

74
Figure 39. Layers Visualization
The map's default background uses a simple topographic base map, which provides
essential geographical context without overwhelming the wildlife data. However, Users can
change the base map based on preferences through the base map button at the top right (Figure
40).

75
Figure 40. Basemap Options

76
Clicking on a data point reveals a pop-up with detailed information about the wildlife
observation, including species name, observed time, and user uploaded images as shown in
Figure 41. This feature encourages users to dive deeper into individual records.
Figure 41. Pop-up Example

77
The hotspot analysis conducted in this study revealed several areas of significant wildlife
activity, providing critical insights into biodiversity patterns within the region. Using the GetisOrd Gi* statistic, high-density clusters of wildlife observations were identified, indicating
ecological hotspots essential for species conservation. Taking Amphibia as an example in Figure
42. These clusters predominantly occurred in protected areas and along known migration
corridors, highlighting their importance for maintaining biodiversity. Conversely, low-density
clusters corresponded to regions impacted by human activities, such as urban development and
agriculture, emphasizing the need for targeted conservation interventions.
Figure 42. Hotspot Analysis for Amphibia

78
4.2 Web Appearance
The Historical Kenya Wildlife Observation application's web design is crafted to ensure
both usability and aesthetic appeal, striking a balance between functionality and visual
simplicity. As Figure 43, the design approach emphasizes an intuitive layout that caters to a
broad audience that may include wildlife enthusiasts, researchers, and conservationists.
Figure 43. Web Appearance
The web's layout consists of three sections: the header, map, and footer. The header
contains the title, additional resources links, and a question mark icon that opens a help pop-up
when clicked. The title is centrally aligned for clarity, while the additional resources links
provide quick access to external wildlife-related websites such as animal protection services and
anti-poaching initiatives.

79
The map section occupies the bulk of the page to maximize the focus on the wildlife data.
This prominent, central feature allows users to interact with the data immediately. Beneath the
map, a "New Observation" button provides access to the Esri Survey123 form for submitting
new wildlife observations.
The footer contains a short disclaimer regarding the source of the data and provides
copyright information. It remains fixed at the bottom of the screen to ensure accessibility on all
pages.
The application employs CSS Flexbox and media queries to ensure the web layout
adjusts to different screen sizes. The map and additional resources components are displayed
horizontally for larger screens, while on smaller screens (such as mobile devices, Figure 44),
elements stack vertically to improve readability and usability. The additional resources links are
moved into a collapsible menu on mobile devices to maximize space for the interactive map. The
map dynamically resizes to fit the available screen, ensuring a consistent experience across all
devices.

80
Figure 44. Wildlife Observation App Mobile View
Pop-up windows are used for the welcome message and the "How It Works" instructional
video (Figure 45 and 46) on both mobile app and larger screens. These pop-ups are designed to

81
be non-intrusive, using smooth transitions and consistent color schemes to enhance the user
experience without interrupting the workflow.
Figure 45. Welcome Message

82
Figure 46. Instruction Video

83
4.3 Data Collection Survey
The data collection survey within the Historical Kenya Wildlife Observation application
plays a pivotal role in gathering new wildlife observations from users. Built using Esri
Survey123, the survey allows users to report their observations in real-time when connected to
the internet, contributing to the growing repository of data used for wildlife conservation efforts.
The user-friendly design ensures that individuals with varying knowledge can easily
submit their wildlife observations. This design follows a straightforward structure. The guides in
each question aid the users in navigating. Key fields in the survey include the species observed,
the date and time of the observation, the location, and the ability to upload photographs (Figure
47).

84
Figure 47. New Observation Form
The survey's visual design is integrated seamlessly into the overall application's aesthetic,
using the same green and earth-tone color scheme as the rest of the web app. This consistency
fosters a unified user experience and reinforces the app's conservation-focused theme.

85
Once the form is submitted, a thank-you message is displayed, and the data is
automatically uploaded to an ArcGIS Online layer. This layer is set to be private due to the
protection of wildlife, and the data will be updated for the application every five years (Figure
48).
Figure 48. Layer Contains New Observed Points (Not Open to Public)
The results presented in this chapter highlight the effectiveness of the Historical Kenya
Wildlife Observation Web Application in visualizing long-term wildlife data and identifying key
spatial patterns. Integrating VGI from iNaturalist with GIS-based spatial analysis tools provides

86
valuable insights into species distributions and ecological trends. Through interactive mapping
and analytical functionalities, the platform enhances accessibility to wildlife data while
supporting conservation decision-making.
However, several challenges and limitations emerged during the implementation process,
including data accuracy concerns, visualization constraints, and ethical considerations regarding
the public sharing of wildlife locations. These factors emphasize the need for continuous
refinement of data integration strategies, symbology improvements, and ethical safeguards to
protect sensitive species data from potential misuse.
Despite these limitations, this study demonstrates the potential of web-based GIS
applications in biodiversity monitoring. The findings from this chapter provide the foundation
for further discussion in Chapter 5, where the implications of these results will be examined in
greater depth, along with recommendations for future improvements and potential expansion
areas.

87
Chapter 5. Conclusions
This chapter delves into the distinctive outcomes of the Historical Kenya Wildlife Observation
project, highlighting its success in creating an innovative platform that leverages communitycontributed data to support wildlife conservation in Kenya. Section 5.1 revisits the project’s
unique objectives, such as integrating historical wildlife data with modern GIS technologies to
produce an interactive and educational tool accessible to conservationists and the public. Section
5.2 focuses on the challenges encountered, including addressing data inaccuracies from VGI,
overcoming ArcGIS platform limitations, and designing a responsive interface that balances
functionality with user-friendliness. Finally, Section 5.3 outlines targeted areas for future
improvement, such as incorporating real-time data streams, enhancing analytical capabilities, and
fostering collaborations with conservation organizations to broaden the tool’s impact. These
discussions underscore the project’s innovative approach to bridging technology, community
involvement, and conservation science.
5.1 App Summary
The Historical Kenya Wildlife Observation project was developed to provide a userfriendly, interactive platform for historical wildlife observation and community-driven data
collection. The main objective was to create a web-based application using ArcGIS tools that

88
would allow users to access historical wildlife observations in Kenya, report new observations,
and contribute to conservation efforts.
The project focused on integrating VGI data from iNaturalist, specifically targeting four
taxonomic groups: Amphibians, Mammals, Aves, and Reptiles. Through a combination of
ArcGIS Pro, ArcGIS Online, and ArcGIS Experience Builder, the application provided an
intuitive interface for users to explore historical data from 2010 to 2020. Survey123 provided
functions for new observation reporting. Collaborating with a responsive HTML web frame
design ensured the platform was accessible across various devices.
The project successfully combined spatial data management, interactive mapping, and
user engagement to promote public awareness of wildlife conservation. This web application
identified all the major national parks and reserves within the study area, which helps users
understand the potential causes of wildlife distribution. Temporal analysis features allow users to
monitor seasonal and long-term changes in species distribution. Users can explore more
information regarding Kenya's wildlife and conservation through the website linked in the web
application. The 5-years data refresh rate helps to avoid illegal activities such as poaching.
However, the Historical Kenya Wildlife Observation Web Application is designed
exclusively for this project. For access, please contact the project owner directly.

89
5.2 Historical Kenya Wildlife Observation Web Application VS. iNaturalist
5.2.1 Data Focus
iNaturalist is primarily designed to collect real-time observations from all over the world.
In addition to wildlife, it also collects data on vegetation and fungus. The Historical Kenya
Wildlife Observation Web Application is designed to focus on the four chosen taxa in Kenya. It
will only present data from 2010-2020 and will only be updated every five years.
5.2.2 Spatial Analysis
In the Historical Kenya Wildlife Observation Web Application, critical wildlife habitats
are highlighted with different colors to guide users with the study area. Spatial analysis tools,
such as hotspot analysis and temporal clustering are employed to help users understand the
wildlife distribution better. A time slide is also added to help users focus data on a specific time
range. However, iNaturalist does not provide any spatial analysis tools, instead offering raw
observation data that can be exported for external analysis.
5.2.3 Map Accessibility
In the Historical Kenya Wildlife Observation Web Application, accessibility features
have been integrated to enhance user experience and ensure inclusivity. Different icons were
applied to represent each taxon—Amphibians, Mammals, Aves, and Reptiles—making it easier

90
for users to distinguish between species groups immediately. To further assist users in navigating
the map, a legend with detailed information about the symbology and layers is prominently
displayed. Additionally, an instruction video is provided, offering a step-by-step guide on how to
browse and utilize the web application effectively.
In contrast, iNaturalist employs a simpler approach to map symbology, using uniform
points to represent all observations. While this simplicity makes the map less cluttered, it can
cause confusion when users attempt to locate or analyze specific taxon data. Moreover,
iNaturalist only provides several guidance links that composed by text and pictures to help users,
which can be difficult to new users.
5.3 Challenges and Limitations
The development of the Historical Kenya Wildlife Observation project encountered
several challenges related to data integration, software constraints, and user accessibility. This
section provides a detailed overview of these challenges and discusses the solutions implemented
during the project.
5.3.1 Data Integration and Cleaning
Integrating VGI from iNaturalist introduced challenges such as inaccurate geolocations,
incorrect species identifications, and missing metadata. Addressing these issues required
extensive data-cleaning processes in ArcGIS Pro. Critical tasks included removing duplicate

91
records and standardizing metadata fields, enabling local geodatabase integration. ArcGIS Pro
tools were developed to identify and flag duplicate entries, reducing manual workload and
improving dataset accuracy. Misplaced data points were corrected by cross-referencing with
authoritative geographic layers, ensuring higher data integrity.
5.3.2 Software Limitations in ArcGIS Suite
During the project, customization constraints within ArcGIS Experience Builder (patch
10.9) presented challenges. The platform's built-in functionalities limited the ability to tailor the
interface fully to the project's requirements. Custom CSS and JavaScript were employed to
overcome these limitations, enhancing the design and expanding the platform's capabilities.
Integration of Survey123 provided flexible data entry forms, allowing users to submit new
observations easily.
Another significant challenge was optimizing performance, especially when handling
large datasets on ArcGIS Online. The project initially faced performance bottlenecks, including
slower loading times and occasional lags. These issues were mitigated by implementing data
filters to limit the number of visible features at any given time and setting visibility ranges to
control the display of layers based on zoom levels. These measures significantly enhanced the
application's responsiveness and ensured a smoother user experience.

92
5.3.3 User Accessibility and Interface Design
Ensuring a responsive and intuitive user interface was a core objective of the project. To
achieve this, a layout that adapted seamlessly across various devices, including desktops, tablets,
and mobile phones, was designed. CSS Flexbox was utilized to ensure that elements resized and
repositioned effectively, providing a consistent user experience regardless of screen size. This
responsive design helped maintain the platform's accessibility, enabling users to interact with the
application without disruption across different devices.
A key focus was also on making the platform simple and user-friendly. Dropdown
menus, tooltips, and clear call-to-action buttons were incorporated to facilitate easy navigation.
Additionally, a help icon linked to an instructional video provided users with guidance on
navigating the platform and using its features. This feature can be beneficial for first-time users,
enhancing their experience by offering step-by-step explanations.
These solutions collectively addressed the primary challenges encountered during the
project, ensuring that the platform remained functional, accessible, and user-friendly. By
overcoming these obstacles, the project created a robust tool for tracking and reporting wildlife
observations, thus supporting broader conservation efforts.

93
5.3.4 Potential Competitors
During this research, Esri and iNaturalist released a beta version of a wildlife tracking
map in June 2024. While this tool was not available during the early phases of the project, it
highlights the rapid advancements in GIS and VGI technologies for conservation. However, the
Historical Kenya Wildlife Observation web application does not aim for profitable goals and
primarily serves as this project's product; this development affirms the necessity of this work and
its focus on addressing region-specific needs, such as historical data integration and enhanced
visualization. Future iterations of this project could explore synergies with these emerging tools,
leveraging their capabilities to enhance biodiversity monitoring and public engagement further.
5.4 Future Work
The Historical Kenya Wildlife Observation project successfully established a platform for
tracking and reporting wildlife observations across Kenya. However, there remains significant
scope for future enhancements and expansions. This section outlines potential future work that
could build upon the current framework to improve the platform's functionality, usability, and
impact further.
5.4.1 Enhanced Data Integration
Future work could focus on expanding the types and sources of data integrated into the
platform. Currently, the platform primarily relies on VGI data from iNaturalist, but incorporating

94
real-time data streams from the iNaturalist API could significantly enhance functionality. This
API allows access to live biodiversity records, enabling automatic updates and providing the
most recent observations for more dynamic conservation planning (iNaturalist API 2025;
iNaturalist GitHub 2025). Additionally, integrating data from other platforms, such as eBird
(eBird) or automated wildlife tracking systems, would enrich the dataset. Advanced technologies
like GPS collars or acoustic sensors could further improve wildlife monitoring capabilities.
However, strict rules and methods are required to avoid poaching.
5.4.2 Advanced Data Analytics and Visualization
While the current platform offers basic map-based visualizations, future enhancements
could include more sophisticated data analysis and visualization tools. For instance, heat maps,
trend analyses, and species distribution models could provide deeper insights into wildlife
behavior and movement patterns. Additionally, the platform could implement AI-driven
predictive analytics to forecast potential changes in species distribution based on historical data
and environmental factors. This would be especially useful for conservation planning and
resource allocation.
5.4.3 Improved User Experience and Interface Design
Despite efforts to make the platform responsive and user-friendly, there are still areas
where the user interface could be further refined. Future work could explore the development of

95
more interactive map elements, such as how weather affect the distribution of wildlfie. Enhanced
filtering options would allow users to customize their viewing experience, making focusing on
specific species or regions of interest easier. Moreover, multilingual support could broaden the
platform's accessibility, enabling users from different linguistic backgrounds to engage with the
tool more effectively.
5.4.4 Enhanced Data Security and Privacy
Given the sensitivity of some wildlife data, especially concerning endangered species,
future work should focus on enhancing data security and privacy. Implementing more robust
encryption protocols and secure data access controls could prevent the unauthorized use of
sensitive information. Additionally, further establishing a system for anonymizing location data
could provide better privacy protections, ensuring that the platform remains compliant with
ethical standards for wildlife conservation and data sharing.
5.4.5 Expanded Educational and Outreach Features
The platform's potential as an educational tool for raising awareness about wildlife
conservation could be further developed. Future versions could incorporate more educational
content, such as interactive tutorials, species fact sheets, and information on conservation
practices. Collaborations with schools and conservation organizations could lead to the

96
development of learning modules, quizzes, or even live-streamed events, enhancing public
engagement and understanding of wildlife conservation issues.
5.4.6 Broader Geographic Expansion
Future iterations of the platform aim to improve its capabilities and expand its geographic
scope beyond Kenya. By adapting to include diverse datasets and addressing region-specific
challenges, the platform could evolve into a comprehensive wildlife monitoring tool for Africa or
even globally. Collaborations with international conservation organizations could facilitate datasharing agreements and create a broader network of contributors, making the platform a go-to
resource for wildlife data worldwide.
Addressing these areas of future work can make the Historical Kenya Wildlife
Observation project a more robust and versatile tool. Continuous development will enable it to
play a more significant role in global wildlife conservation, supporting efforts to understand and
protect biodiversity on a larger scale.

97
References
Acharyya, Nirupam, and Jatisankar Bandyopadhyay. "Remote Sensing and GIS for Wildlife
Management." In National Level Seminar on Defaunation and Conservation; Tucson
Herpetological Society: Tucson, AZ, USA, pp. 65–71. 2017
AgriLife Today. “A New Way to Map With Conservation Data.” AgriLife Today, May 2, 2024.
https://agrilifetoday.tamu.edu/2021/10/20/a-new-way-to-map-with-conservation-data.
Alesheikh, AA., Hossein Helali and HA. Behroz. “Web GIS: Technologies and Its
Applications.” (2002).
Ames, Daniel & Michaelis, Christopher & Dunsford, Ted. (2007). Introducing the MapWindow
GIS Project. 2. 13-16.
Anderson, Bradley, and Johan Jooste. "Wildlife Poaching: Africa's Surging Trafficking
Threat." Africa Security Briefs 28 (2014).
Ballantyne, Roy, and Jan Packer. 2009. “Introducing a Fifth Pedagogy: Experience‐based
Strategies for Facilitating Learning in Natural Environments.” Environmental Education
Research 15 (2): 243–62.
Beaird, Jason, Alex Walker, and James George. The principles of beautiful web design. SitePoint
Pty Ltd, 2020.
“Better Species Mapping Can Improve Conservation Efforts.” ScienceDaily, April 18, 2018.
https://www.sciencedaily.com/releases/2018/04/180418144732.htm.
Butchart, Stuart HM, Matt Walpole, Ben Collen, Arco Van Strien, Jörn PW Scharlemann,
Rosamunde EA Almond, Jonathan EM Baillie et al. "Global biodiversity: indicators of
recent declines." Science 328, no. 5982 (2010): 1164-1168.
Channell, Rob, and Mark V. Lomolino. "Dynamic biogeography and conservation of endangered
species." Nature 403, no. 6765 (2000): 84-86.
Chakraborty, Debasish, Debanjan Sarkar, Shubham Agarwal, Dibyendu Dutta, and Jaswant R.
Sharma. "Web based GIS application using open source software for sharing geospatial
data." international journal of advanced remote sensing and gis 4, no. 1 (2015): 1224-1228.

98
Chen, Yu-Horng. “Development of the Animal Conservation Digital Learning AIDS and
Assessments through the Industry-University Collaborative Course.” MDPI, July 6, 2021.
https://www.mdpi.com/2071-1050/13/14/7524.
Duckett, Jon. HTML & CSS: design and build websites. Vol. 15. Indianapolis, IN, USA:: Wiley,
2011.
Dwyer, Ross G., Charles Brooking, Wilfred Brimblecombe, Hamish A. Campbell, Jane Hunter,
Matthew Watts, and Craig E. Franklin. "An open Web-based system for the analysis and
sharing of animal tracking data." Animal biotelemetry 3 (2015): 1-11.
eBird. 2025. Accessed January 25, https://ebird.org/home
Eliason, Stephen L. "The illegal taking of wildlife: Toward a theoretical understanding of
poaching." Human Dimensions of Wildlife 4, no. 2 (1999): 27-39.
Elwood, Sarah, Michael F. Goodchild, and Daniel Z. Sui. "Researching Volunteered Geographic
Information: Spatial Data, Geographic Research, and New Social Practice." Annals of the
Association of American Geographers 102, no. 3 (2012): 571-590.
Esri. "Beta Wildlife Tracking Map." Accessed June 15, 2024. https://www.esri.com
Flanagin, Andrew J. and Miriam J. Metzger. “Digital Media and Youth: Unparalleled
Opportunity and Unprecedented Responsibility.” (2008).
Fletcher, Robert, and M. Fortin. Spatial ecology and conservation modeling. Cham: Springer
International Publishing, 2018.
Fryxell, John M., Anthony R. E. Sinclair, and Graeme Caughley. Wildlife Ecology,
Conservation, and Management. Hoboken, NJ: John Wiley & Sons, 2014
Fu, Pinde. "Getting to know Web GIS." Photogrammetric engineering & remote SenSing 84, no.
2 (2018): 59-60.
Garrah, E., Danby, R.K., Eberhardt, E. et al. Hot Spots and Hot Times: Wildlife Road Mortality
in a Regional Conservation Corridor. Environmental Management 56, 874–889 (2015).
https://link.springer.com/article/10.1007/s00267-015-0566-1
Goodchild, Michael F., and Linna Li. "Assuring the Quality of Volunteered Geographic
Information." Spatial Statistics 1 (2012): 110-120.
https://www.sciencedirect.com/science/article/pii/S2211675312000097?via%3Dihub

99
Haklay, Muki. "Citizen Science and Volunteered Geographic Information: Overview and
Typology of Participation." In Crowdsourcing Geographic Knowledge: Volunteered
Geographic Information (VGI) in Theory and Practice, edited by Daniel Sui, Sarah
Elwood, and Michael Goodchild, 105-122. Dordrecht: Springer, 2013.
Haklay, Muki. How Good is Volunteered Geographical Information? A Comparative Study of
OpenStreetMap and Ordnance Survey Datasets. Environment and Planning B: Planning
and Design, 37(4), 682-703. https://journals.sagepub.com/doi/10.1068/b35097
iNaturalist. A Community for Naturalists. Accessed August 10, 2024. https://www.inaturalist.org
iNaturalist API. 2025. Accessed January 24. https://api.inaturalist.org/v1/docs/
iNaturalist GitHub. 2025. Accessed January 24.
https://github.com/inaturalist/inaturalist/tree/main/public
IUCN Red List of Threatened Species. “The IUCN Red List of Threatened Species,” n.d.
https://www.iucnredlist.org/.
LaZerte, Stefanie E., Matthew W. Reudink, Ken A. Otter, Jackson Kusack, Jacob M. Bailey,
Austin Woolverton, Mark Paetkau, Adriaan de Jong, and David J. Hill. "feedr and
animalnexus. ca: A paired R package and user‐friendly Web application for transforming
and visualizing animal movement data from static stations." Ecology and evolution 7, no.
19 (2017): 7884-7896.
Lehtomäki, Joona, Erkki Tomppo, Panu Kuokkanen, Ilkka Hanski, and Atte Moilanen.
"Applying spatial conservation prioritization software and high-resolution GIS data to a
national-scale study in forest conservation." Forest Ecology and Management 258, no. 11
(2009): 2439-2449.
Liu, Yeqiang, Weiran Li, Xue Liu, Zhenbo Li, and Jun Yue. “Deep Learning in Multiple Animal
Tracking: A Survey.” Computers and Electronics in Agriculture 224 (June 25, 2024):
109161.
https://www.sciencedirect.com/science/article/pii/S0168169924005520?via%3Dihub
Lorini, Maria Lucia, Adriana Paese, and Alexandre Uezu. "GIS and spatial analysis meet
conservation: a promising synergy to address biodiversity issues." Natureza &
Conservação 9, no. 2 (2011): 129-144.
“Mapping for Conservation.” Natural Reserve System, June 12, 2019. https://ucnrs.org/mappingfor-conservation/.

100
Martin, Michael & Leszczynski, Agnieszka & Gaillard, J.C.. (2022). Maps, volunteered
geographic information (VGI) and the spatio-discursive construction of nature. Digital
Geography and Society. 3. 100029. 10.1016/j.diggeo.2022.100029.
Martin, Stuart. “The Africa GeoPortal - Geospatial Content and Tools for Africa (Open Data Day
2021).” The Africa GeoPorta, March 4, 2021.
https://zivahub.uct.ac.za/articles/presentation/The_Africa_GeoPortal_-
_Geospatial_Content_and_Tools_for_Africa_Open_Data_Day_2021_/14161304/1
Menon, Shaily, and Kamaljit S. Bawa. "Applications of geographic information systems, remotesensing, and a landscape ecology approach to biodiversity conservation in the Western
Ghats." Current science (1997): 134-145.
Nowak, Maciej M., Katarzyna Dziób, Łukasz Ludwisiak, and Julian Chmiel. "Mobile GIS
applications for environmental field surveys: A state of the art." Global Ecology and
Conservation 23 (2020): e01089.
Pedersen, Tore, Anne-Mari Jordal, Hanne Monke Bernhoft, and Per Stabell. "Implementation of
a GIS-based Management Tool for Conservation of Biodiversity within the Municipality of
Oslo, Norway." Landscape and Urban Planning 63, no. 3 (2003): 429-438.
https://www.sciencedirect.com/science/article/pii/S0169204603001488
Polly. “National Parks and National Reserves in Kenya | Kenya Wildlife Safaris.” Kenya
Wildlife Tours, October 28, 2021. https://www.kenyawildlifetours.com/national-parksand-national-reserves-in-kenya/.
Prasad, Nupoor, Manoj Semwal, and Parth Sarathi Roy. "Remote Sensing and GIS for
Biodiversity Conservation." In Recent Advances in Lichenology: Modern Methods and
Approaches in Biomonitoring and Bioprospection, Volume 1, 151–79. Springer, 2015
Rao, Mahesh & Fan, Guoliang & Thomas, Johnson & Cherian, Ginto & Chudiwale, Varun &
Awawdeh, Muheeb. (2007). A web-based GIS Decision Support System for managing and
planning USDA's Conservation Reserve Program (CRP). Environmental Modelling &
Software. 22. 1270-1280. 10.1016/j.envsoft.2006.08.003.
Sheldrick Wildlife Trust. "Anti-Poaching Efforts." Accessed August 2024.
https://www.sheldrickwildlifetrust.org
Silvestro, Daniele, Stefano Goria, Thomas Sterner, and Alexandre Antonelli. “Improving
Biodiversity Protection through Artificial Intelligence.” Nature News, March 24, 2022.
https://www.nature.com/articles/s41893-022-00851-6.

101
Siriba, David N., and Sagi Dalyot. "Adoption of volunteered geographic information into the
formal land administration system in Kenya." Land Use Policy 63 (2017): 279-287.
Stefan Steiniger, Geoffrey J. Hay, Free and open source geographic information tools for
landscape ecology, Ecological Informatics, Volume 4, Issue 4, 2009, Pages 183-195, ISSN
1574-9541, https://doi.org/10.1016/j.ecoinf.2009.07.004.
Steiniger, Stefan & Hunter, Andrew. (2013). The 2012 Free and Open Source GIS Software Map
– A Guide to facilitate Research, Development and Adoption. Computers, Environment
and Urban Systems. 39. 10.1016/j.compenvurbsys.2012.10.003.
Tuia, Devis, Benjamin Kellenberger, Sara Beery, Blair R. Costelloe, Silvia Zuffi , Benjamin
Risse, Alexander Mathis, et al. “Perspectives in Machine Learning for Wildlife
Conservation.” Nature News, February 9, 2022. https://www.nature.com/articles/s41467-
022-27980-y
Van Dooren, Thom, Catherine J. Price, Peter B. Banks, Oded Berger-Tal, Matthew Chrulew,
Jane Johnson, Gabrielle Lajeunesse, Kate E. Lynch, Clare McArthur, Finn C.G. Parker,
Myles Oakey, Benjamin J. Pitcher, Colleen Cassady St. Clair, Georgia Ward-Fear, Sam
Widin, Bob B.M. Wong, and Daniel T. Blumstein. "The Ethics of Intervening in Animal
Behaviour for Conservation." Trends in Ecology & Evolution 38, no. 9 (2023): 822-830.
https://www.cell.com/trends/ecology-evolution/fulltext/S0169-5347(23)00091-
5?_returnURL=https%3A%2F%2Flinkinghub.elsevier.com%2Fretrieve%2Fpii%2FS0169
534723000915%3Fshowall%3Dtrue
Valletta, John Joseph, Colin Torney, Michael Kings, Alex Thornton, and Joah Madden.
“Applications of Machine Learning in Animal Behaviour Studies.” Animal Behaviour,
January 23, 2017. https://www.sciencedirect.com/science/article/pii/S0003347216303360.
Verma, Jyoti, and Hemlata Pant. “Wildlife Conservation and Management Using GIS and
Remote Sensing.” Research Gate, September 2022.
https://www.researchgate.net/publication/363263117_Wildlife_Conservation_and_Manage
ment_using_GIS_and_Remote_Sensing.
Watson, James EM, Nigel Dudley, Daniel B. Segan, and Marc Hockings. "The performance and
potential of protected areas." Nature 515, no. 7525 (2014): 67-73.
Zharikov, Yuri, Greg A. Skilleter, Neil R. Loneragan, Thomas Taranto, and Bronwyn E.
Cameron. “Mapping and Characterising Subtropical Estuarine Landscapes Using Aerial
Photography and GIS for Potential Application in Wildlife Conservation and

102
Management.” Biological Conservation, 2005.
https://www.sciencedirect.com/science/article/pii/S0006320705001369?casa_token=qxuic
wZ2aj4AAAAA:qYkPKrs8tZ_LN-C1--
EUowZC1fh7JSw5M8uJiDUF1CpC9QtsDLL8STScJ9UF9uwcgI5xXpyhJh0.

Abstract (if available)

Abstract The conservation of wildlife is a global imperative, driven by the need to preserve biodiversity and maintain the delicate balance of ecosystems. In Africa, a continent renowned for its rich and diverse wildlife, these efforts are particularly critical. However, the challenges posed by habitat loss, climate change, and illegal poaching have intensified, necessitating innovative approaches to wildlife conservation. In response to these challenges, this thesis presents the development and implementation of the Historical African Wildlife Siting App, a Geographic Information System (GIS)-based platform designed to track and report wildlife sightings across Africa. Results of this thesis show the technical development of the Historical Kenya Wildlife Siting App. This platform integrated GIS and VGI to track historical wildlife observations in Kenya, cooperated with user interface design techniques and data management strategies. It also examines the app's potential impact on wildlife conservation, highlighting the role of digital platforms in fostering community engagement and improvements may applied in the future.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Tracking trends in earthquakes and tropical storms: a web GIS application

PDF

Trojan Food Finder: a web-based GIS campus food sharing application

PDF

Web GIS as a disease management workspace: enabling advocacy at multiple scales across multiple continents with the case of tungiasis

PDF

Happy Traveler: discovering joy on university campuses and beyond through a web-based GIS application

PDF

Implementing spatial thinking with Web GIS in the non-profit sector: a case study of ArcGIS Online in the Pacific Symphony

PDF

Geospatial web application development to access irrigation asset data: Veterans Affairs Palo Alto Health Care System

PDF

Student engagement capabilities in mobile GIS: A framework for mobile GIS education

PDF

Geographic information systems and marketing: a transdisciplinary approach to curriculum development

PDF

Exploring global natural disaster and climate migration data: a Web GIS application

PDF

A Web GIS application for airport pavement management

PDF

Building a spatial database of biochar research and practice with Web-GIS

PDF

Spatial narrative of the invasive lionfish in the Western Atlantic and Caribbean oceans: a GIS story map

PDF

Cartographic design and interaction: An integrated user-centered agile software development framework for Web GIS applications

PDF

Visualizing historic space through the integration of geographic information science in secondary school curriculums: a comparison of static versus dynamic methods

PDF

Development of a Web GIS application to aid marathon runners in the race selection and planning process

PDF

Bringing GIS to a small community water system

PDF

Applying GIS to landscape irrigation systems: a case study of the Music Academy of the West campus in Montecito, CA

PDF

Rails to trails web mapping application for the great Redwood trails: mapping northern California’s repurposed trails

PDF

Building a spatial database for agricultural record keeping and management on a regenerative farm

PDF

Analysis of park accessibility in Redan, Georgia Web GIS application

Asset Metadata

Creator Lin, Tianle (author)

Core Title Historical observations of wildlife in Kenya: a Web GIS application

School College of Letters, Arts and Sciences

Degree Master of Science

Degree Program Geographic Information Science and Technology

Degree Conferral Date 2025-05

Publication Date 02/03/2025

Defense Date 12/18/2024

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

Tag Africa,Esri,GIS,OAI-PMH Harvest,web application,web map,Wildlife

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Ruddell, Darren (committee chair), Loyola, Laura (committee member), Swift, Jennifer (committee member)

Creator Email tianle@usc.edu

Unique identifier UC11399GBJG

Identifier etd-LinTianle-13809.pdf (filename)

Legacy Identifier etd-LinTianle-13809

Document Type Thesis

Format theses (aat)

Rights Lin, Tianle

Internet Media Type application/pdf

Type texts

Source 20250204-usctheses-batch-1240 (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