/

University of Southern California Dissertations and Theses

/

Scenario-based site suitability analysis and framework for biodiversity conservation: agricultural zone, Galapagos Archipelago, Ecuador

(USC Thesis Other)

Scenario-based site suitability analysis and framework for biodiversity conservation: agricultural zone, Galapagos Archipelago, Ecuador

PDF

Transcript (if available)

Content Scenario-Based Site Suitability Analysis and Framework for Biodiversity Conservation:
Agricultural Zone, Galapagos Archipelago, Ecuador
by
Petros Maskal
A Thesis Presented to the
FACULTY OF THE USC COLLEGE OF LETTERS, ARTS AND SCIENCE
University of Southern California
In Partial Fullfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(GEOGRAPHIC INFORMATION SCIENCE AND TECHNOLOGY)
August 2021
Copyright 2021 Petros Maskal
ii

For Nebabie

iii

Acknowledgements

I want to thank my thesis committee at the University of Southern California, Dr. Darren M.
Ruddell, for his continuous guidance and advice at each stage of this thesis and to Dr. Laura C.
Loyola, and Dr. Leilei Duan for their knowledge, advice, and geographic insight. I also want to
thank the professors of the Spatial Sciences Institute: Dr. John Wilson, Dr. Robert O. Vos, Dr.
Jennifer N. Swift, Dr. Jennifer M. Bernstein, Dr. Andrew J. Marx, Dr. Vanessa Griffith Osborne,
and Dr. Katsuhiko Oda for their instruction and guidance.

iv

Table of Contents

Dedication ....................................................................................................................................... ii
Acknowledgements ........................................................................................................................ iii
List of Tables ................................................................................................................................. vi
List of Figures ............................................................................................................................... vii
List of Abbreviations ..................................................................................................................... ix
Abstract ........................................................................................................................................... x
CHAPTER 1: Introduction ............................................................................................................. 1
1.1 The Need to Reduce Biodiversity Loss ...............................................................................1
1.1.1 Social Context of Biodiversity Loss ...........................................................................2
1.1.2 Biodiversity Loss Situation in the Galapagos Arichelago ..........................................3
1.2 Goal of this Study ................................................................................................................5
1.3 Scope of Framework ............................................................................................................7
1.4 The Study Workflow............................................................................................................8
CHAPTER 2: Relevant Background Research ............................................................................ 10
2.1 Review of Relevant Studies ...............................................................................................10
2.2 Site Selection for Scenario-Based Land Evaluation ..........................................................11
2.3 Methods for Site Suitability Analysis ................................................................................12
CHAPTER 3: Scenario-Based Land Evaluation Framework for Biodiveristy Conservation in the
Galapagos Archipelago ................................................................................................................ 16
3.1 Study Area .........................................................................................................................16
3.2 Data ....................................................................................................................................17
3.3 Methodology ......................................................................................................................21
3.3.1 Fuzzy Overlay Process ..............................................................................................21
3.3.2 Land Evaluation Scenarios .......................................................................................23
3.3.3 Ground Truthing .......................................................................................................23
3.3.4 Suitability Criteria for the General Framework ........................................................27
3.4 Major Model Criteria and the Implied Constraints ............................................................34
v

3.5 Minor Model Criteria ..........................................................................................................48
CHAPTER 4: Scenario-Based Suitability Framework Implementation ...................................... 56
4.1 Scenario 1: Agroforestry Production .................................................................................56
4.1.1 Agroforestry Production Sub Model Results ............................................................56
4.1.2 Agroforestry Production Scenario ............................................................................62
4.2 Scenario 2: Tourism Development ....................................................................................64
4.2.1 Tourism Development Sub Model Results ...............................................................65
4.2.2 Tourism Development Scenario ...............................................................................67
4.3 Interrelationship of Scenarios ............................................................................................70
CHAPTER 5: Summary and Conclusion ...................................................................................... 71
5.1 Assessment of Model ........................................................................................................ 71
5.2 Future Work ...................................................................................................................... 71
5.3 Applicability of Research ................................................................................................. 73
References ..................................................................................................................................... 75
Appendices .................................................................................................................................... 80
Appendix A Analysis of the scenario one hydrology category of criteria factors .................. 80
Appendix B Analysis of the scenario one topographic category of criteria factors ............... 83
Appendix C Analysis of the scenario one built environment category of criteria factors ...... 84
Appendix D Analysis of the scenario one temperature category of criteria ........................... 85
Appendix E Analysis of the scenario one soil category of criteria ......................................... 86
Appendix F Analysis of the scenario one aspect category of criteria ..................................... 87
Appendix G Analysis of the scenario two built environment category of criteria factors ..... 88
Appendix H Analysis of the scenario two temperature category of criteria ........................... 91

vi

List of Tables

Table 1 The site suitability classification system………………………………………………..14

Table 2 Summary of Data ……………………………………………………………………….18

Table 3 Scenario 1 Site Suitability Analysis Criteria …………………………………………...28

Table 4 Scenario 2 Site Suitability Analysis Criteria ……………..……………………….........30

Table 5 Data Summary and Attributes of Analysis Criteria ……………..……………………...31

Table 6 The suitable area measurement and membership class results of scenario one ……...…64

Table 7 The suitable area measurement and membership class results of scenario two ……..…69

Table 8 The defuzzification percentages of suitable area for both scenario one and two ………70

vii

List of Figures

Figure 1 Research Study Area …………………………………………………………………....6

Figure 2 The study workflow diagram …………………………………………………………...8

Figure 3 The Site Suitability Analysis workflow for both scenario 1 and scenario 2……...……22

Figure 4 The site map of ground truth locations recorded across Santa Cruz Island…...……….24

Figure 5 The site map of ground truth locations recorded across Isabela Island ……….……….25

Figure 6 The site map of ground truth locations recorded across San Cristobal Island ……..….26

Figure 7 The visualization of average annual precipitation criteria ………………...…………..35

Figure 8 The visualization of the terrain elevation criteria ……………………………..….……37

Figure 9 The visualization of the potential evapotranspiration criteria ……………….………...39

Figure 10 The visualization of the potential aridity criteria……………………………..….…...40

Figure 11 The criteria and visualization of percent slope of terrain …………………...………..42

Figure 12 The visualization of the market center criteria ………………………...…….…….....43

Figure 13 The visualization of the primary and secondary road network criteria ………………45

Figure 14 The visualization of points of access to coastal port routes criteria ………….………46

Figure 15 The visualization of land use land cover classified criteria …………………...……...48

Figure 16 The visualization and criteria of the soil drainage classes …………………….……..50

Figure 17 The visualization of the average annual temperature criteria ……………..………….53

Figure 18 The visualization of the aspect criteria …………………………………………...…..55

Figure 19 Map of fuzzy membership for criterion 1a, 1b and 1c .…………………...…...……..57

Figure 20 Map of fuzzy membership for criterion 2a and 2b ……………………….…………..58

Figure 21 Map of fuzzy membership for criterion 3a and 3b ………………………….………..59

Figure 22 Map of fuzzy membership for criterion 4a ………………………….………………..60

viii

Figure 23 Map of fuzzy membership for criterion 5a ……………………….…………………..60

Figure 24 Map of fuzzy membership for criterion 6a ………………………………….………..61

Figure 25 The fuzzy overlay suitability result of scenario 1-agroforestry production……….….63
.
Figure 26 Map of fuzzy membership for criterion 3c, 3d, 3e and 3f ……………..……………..65

Figure 27 Map of fuzzy membership for criterion 4b ………………………………….………..66

Figure 28 The fuzzy overlay suitability result of scenario 2-tourism development ……….……68

ix

List of Abbreviations

BV Bella Vista

CGIAR Consultative Group on International Agricultural Research

CMt Camote Volcano

CrM Crocker Mountain

CSI Consortium for Spatial Information

GIS Geographic Information Systems

GNP Galapagos National Park

FAO Food and Agricultural Organization

ISRIC International Soil Reference and Information Centre

ITCZ Inter-Tropical Convergence Zone

LULC Land Use Land Cover

MDGs Millennium Development Goals

NGOs Non-Governmental Organization

OC El Occidente

PAy Puerto Ayora

PET Potential Evapotranspiration

Sr Santa Rosa

SSA Site Suitability Analysis

USC University of Southern California

USDA United States Department of Agriculture

UN United Nations

UNESCO United Nations Educational, Scientific and Cultural Organization

x

Abstract

Galapagos Island’s current agricultural system of monocropping, massive food imports, and
a booming tourism sector has provided an increase in income for most galapaguenos that reside
in the island but has been deemed unsustainable by the UNESCO World Heritage Organization.
The tourism-driven urban development and monoculture system of food production have
contributed to declines in water, wildlife habitat, soil quality, and an overall loss in biodiversity.
This tourism sector growth along with a reduction in agroforestry production has reduced the
income diversification potential for galapaguenos that reside in the islands and continues to
threaten biodiversity. The most notable and critical of these global initiatives around biodiversity
is goal seven of the global Millennium Development Goals (MDGs) of the United Nations (UN)
which has since been translated into goal fifteen of the revised Sustainable Development Goals
(SDGs). The goal seven of MDGs was targeted at ensuring environmental sustainability and
parts of this goal were eventually folded into goal fifteen of the SDGs targeted at restoration and
promotion of sustainable use of terrestrial ecosystems or life on land. These targets for goal
fifteen have yet to be achieved. The scenario-based fuzzy modeling study was designed to
support organizations focused on land use planning and management for agroforestry production
and tourism development within the Galapagos Islands agricultural zone of San Cristobal, Santa
Cruz, and Isabela utilizing a site suitability analysis framework. The framework was developed
based on (1) contextual ecosystem requirements, (2) proximity to built environment
infrastructure, and (3) availability of data. The framework implementation identified scenario
1:agroforestry production as being suitable across 20 percent of the study area or 12,386.40 acres
and scenario 2: tourism development being 58 percent suitable within the study area or 35,920.56
acres.
  1

CHAPTER 1: Introduction

1.1 The Need to Reduce Biodiversity Loss
The major global issues of inequitable education, poverty, and lack of basic health services
have been long-lasting across the globe. In the year 2000, the United Nations (UN) created the
Millennium Development Goals (MDGs) which was eventually folded into the revised
Sustainable Development Goals (SDGs) in the year 2015. These development goals were
developed to set targets and take action towards solving these specific humanitarian issues of
poverty, education, children’s health, sustainable environment, economic development, and
disease prevention. These MDGs and SDGs serve as a global framework for development. The
global community has realized progress towards many of the MDGs targets since 2000, and
some of the SDGs since 2015 but many targeted areas still require much attention.
Two of the targets (7A,7B) of the MDGs are to substantially reduce biodiversity loss,
achieved, by 2010 (UN MDG Report 2015) and to reverse the loss of environmental resources
which was translated into the SDGs (15) in the year 2015 targeting the restoration and promotion
of life on land. This MDGs declaration and the UN’s efforts do appear to have been successful in
reducing the loss of rich biodiverse forests. Net loss in forest area declined from 8.3 million
hectares annually in the 1990s to an estimated 5.2 million hectares (an area about the size of
Costa Rica) each year from 2000 to 2010 (UN MDG Report 2015, 1). This successful reduction
in biodiversity and environmental resource loss has been in large part due to the expanded
coverage of protected areas since the 1990s. However, this success in environmental
sustainability has not been universal especially in regions of Africa and South America where
much of the deforestation occurs. Many aspects of the protection and conservation of these
  2

biodiverse hotspots need improvement. These include effective and equitable management and
connectivity, and protection of areas important for biodiversity and ecosystem services,
especially ecologically representative protected area networks (UN MDG Report 2015).
A large portion of these ecologically important species that help maintain the functioning
ecosystem required for environmental resources is endemic to South America with some located
in the Galapagos Islands known as a hotspot of species endemism. Until 2001 a total of 2,289
terrestrial invertebrate species were registered, of which more than half are supposed to be
endemic to the Galapagos Islands (Zachos, 2014).
These conservation and protection efforts in the Galapagos Archipelago and across the globe
are under a race against time to save the critical plant and animal species from extinction. Many
cantons or provinces within the Archipelago are not on track to meet the previous MDGs or
many of the SDGs at the current rate of progress. It is widely recognized that there is a strong
need for more effective planning and better decision-making if the MDGs of the UN is going to
be met (JC Hyneman, 2014). The study focuses on the Galapagos Archipelago given that it is a
globally important region of biodiversity, a hotspot of species endemism, and a source of
environmental resources for its residence.
1.1.1 Social Context of Biodiversity Loss
The effectiveness of the MDGs has been long contested which has led to the formation of
SDGs. Many perspectives exist on this effectiveness of MDGs and whether SDGs are equitable
in its benefits to the private and public organizations of society. This study discusses MDGs and
SDGs as an established international framework and reference point for the scope, need, and
overall benefit of biodiversity loss reduction and environmental resource conservation. The
conversation surrounding MDGs and SDGs is publicly available. A general overview of the
  3

discussions around MDGs can be found on the UN official website and by independent
researchers like Shobha Raghuram, Manasi Kumar, and Erica Burman. The articles provide both
an appreciation and critique of the MDGs (Editorial Critical/Subaltern Perspective on UN MDGs
2009). While articles on SDGs suggest that this revised suit of goals has the potential for holistic
sustainable development but also carries some major limits to change. The article by Regina
Scheyvens and others provides an overview of this discussion on sustainability potential, limits
to change and this movement towards inequitable benefits among the public and private sector
when it comes to driving SDGs (Scheyvens et al., 2016).
This discussion of whether to conserve biodiversity-environmental resources or continue
urban development for economic benefit and at what rate has been one of great contention. The
contentious discussion is taking place globally, regionally, locally, and the Galapagos
Archipelago is no exception. Managers, policy-makers, and conservationists must serve multiple
masters, many of whom have different points of view, expectations, demands, influence, and
power (Epler, 2007). Some of these masters include the global and local scientific community,
the local population, tourist industry, the tourist, nation of Ecuador, and developers. These
challenges that cause contention specifically around biodiversity conservation and environmental
resources will be discussed in the next section. However, most of these challenges especially
around social factors are beyond the scope of this study.
1.1.2 Biodiversity Loss Situation in the Galapagos Archipelago
The Galapagos Archipelago is an autonomous region of Ecuador approximately 1000 km
(600 miles) off the coast of mainland Ecuador, and is considered one of the natural wonders of
the world with renowned natural sites and natural resources. Unlike other oceanic archipelagos,
the ecological and evolutionary processes characteristic of the Galapagos have been minimally
  4

affected by human activities, and the archipelago still retains most of its original, unique
biodiversity. However, several recent reports suggest that the development model has turned
unsustainable and that the unique values of the archipelago might be seriously at risk (González
et al., 2008). The Galapagos was added to the list of UNESCO World Heritage in Danger in
2007 (González et al., 2008). The community of research provides clear evidence of the linkage
between changes in ecology and economics across the Galapagos Islands. An increase in tourism
and a decline in the agricultural and fishing sectors are influencing invasive species and urban
development growth. A tourism sector providing greater income at the cost of resilient
livelihoods and biodiversity loss is unsustainable. This model of tourism reduces the Galapagos
system’s resilience through its effects on population growth, economy, invasive species arrival,
resource consumption, and globalization effects on island residents (González et al., 2008).
The concept of agroforestry is an alternative land management/diversified production
method that can provide economic and ecological benefits as studies in Costa Rica (Ricketts et
al., 2004) and Indonesia (Steffan-Dewenter, I. et al., 2007) suggest. Agroforestry refers to the
land-use systems where woody perennials such as trees, shrubs, palms, bamboo, etc. are
cultivated on the related land units as agricultural crops and/or animal rearing. It has been
practiced in many countries to offer a wide range of economic, social, and ecological benefits by
increasing per capita income of the farms by planting high-value tree species (Ahmad et al.,
2017).
The declining agroforestry sector within the Galapagos Islands is shifting further into a
predominantly mono-cropped food production system with cattle grazing on pasture or fallowing
being the major land use. This monoculture along with tourism development within the
Galapagos agriculture zone has greatly contributed to this biodiversity loss with the spread and
  5

proliferation of invasive species like guava and blackberry. Information now exists that shows
the direct relationship between the abandonment of agricultural land and the subsequent increase
of area affected by invasive plant species (Guzman JC et al., 2013). This agricultural system and
the food production value chain that is associated with its economic efficiency are threatening
the complex diversified and resilient passive agroforestry systems of the Galapagos.
The community of research analyzes several land management scenarios factoring in the
major Galapagos sectors of agriculture, fisheries, and tourism using system dynamics models and
frameworks. Much of the research suggests locally resilient and diversified food systems as
being a viable option to transition to a sustainable Galapagos economy while maintaining and
restoring its biodiversity. Diversification of production systems provides a variety of food
products for personal consumption, local market access and agritourism, which ultimately builds
socioecological resilience (Altieri, 2013).
1.2 Goal of this Study

The goal of the research study was to support organizations focused on land use planning for
agroforestry conservation and tourism development within the Galapagos Islands agricultural
zone of San Cristobal, Santa Cruz, and Isabela (fig.1) by creating a site suitability framework and
demonstrating the method of implementation. The implementation of this framework produced
recommendations of regions that are optimal for agroforestry production in scenario 1 and
tourism development in scenario 2 that minimize risk to biodiversity and maximize the benefit
for the Galapagenous community.

  6

Figure 1. The research study area (agricultural zone) on Isla San Cristobal, Santa Cruz and
Isabela, Ecuador.

The framework was applied to a specific study area and context as a model. In framework
implementation, the model was refined for optimal performance and to meet the needs of a user’s
context. The project has two components per model scenario: (1) the development of the
framework (study), and (2) the implementation of the framework (model fitting).
The model was created in a standard geographic information system (GIS) format utilizing
site suitability analysis mapping techniques that can be applied in environments that lack large
sources of data. This limited availability of data is often the case in developing regions. The
purpose of this framework and modeling was to produce information that can be used to improve
  7

landscape planning and land allocation that contributes to ecologically and economically
balanced sustainable development which in turn conserves biodiversity.
The suitability mapping products and framework can be used by NGOs, local organizations,
and government as a precursor to on-the-ground engineering, soils, or ecological surveys, thus
limiting the area that needs to be assessed in detail while saving resources of time and money.
1.3 Scope of Framework
It is preferred to acquire access to an agroforestry plant database for species-specific habitat
requirements and to factor in the Galapagos Islands autonomous regional census data on tourist
activity when assessing site suitability. However, quality data on agroforestry plant species
endemic to the Galapagos Islands and the Galapagos Islands census data were rare to find or
inaccessible as a complete record-dataset. So instead of developing a site suitability analysis
(SSA) model for a region that is data-rich like Europe with the plan to apply it in a data deficient
region, a limit was placed on the design to exclusively use data readily available to the
Galapagos Archipelago and Ecuadorian public. This was designed to ensure the methods used
are suitable for the region. To meet the goals of regional model versatility it was necessary to not
use agroforestry species-specific data, Ecuadorian census data, or high-resolution commercial
satellite data. Just publicly available datasets with global coverage were selected to ensure the
framework could be replicated. Given these data limitations, the model and framework focused
on locating sites for agroforestry production and tourism development as different scenarios
based on the specific cultural and physical geography instead of using plant species or
demographic-related metrics. Therefore, in order to locate these scenario-based sites that meet
adequate physical and cultural geography requirements, criteria were developed based on (1)
  8

contextual ecosystem requirements, (2) proximity to built environment infrastructure, and (3)
availability of data. The criteria selection details are covered in chapter three.
The framework was designed to be applied throughout the Galapagos Islands and the greater
humid highlands of mainland Ecuador. A study region within the Galapagos Islands was selected
due to its biodiversity significance outlined in previous sections, to assess the effectiveness of the
framework and its general application.
Given the data deficient context of the study region, the effectiveness of the general
framework was accessed by comparing sites identified as suitable (meeting ecosystem or
infrastructure requirements) with the locations of existing remotely sensed land use land cover
(LULC) data for agriculture and urban classes. This assumes that the existing urban and
agricultural land use land cover locations are ideal which is unlikely to be universally correct but
does provide a way of refining the model and measuring output.
1.4 The Study Workflow
A SSA study within a GIS often has similar workflows and this study follows a similar basic
approach, starting with a detailed literature and methods review followed by the appropriate
selection of criteria to be analyzed (fig.2).

Figure 2. The study workflow diagram.
  9

The study continues in chapter two with a review of relevant published literature on the
current biodiversity loss situation in the Galapagos Archipelago, scenario-based land evaluation
modeling, and methods of site suitability analysis. The third chapter covers the study area, data
attributes, methodology selected, the general framework developed, ground-truthing results, and
the details of suitability criteria. The fourth chapter walks through the results of model
implementation in detail by describing the evaluation of results for each modeling factor and the
scenario results. This study concludes with a final chapter discussing the research findings and
areas for future work.

  10

CHAPTER 2: Relevant Background Research
This chapter discusses the literature relevant to the study. The site selection for scenario-
based land evaluation is then outlined. Finally, the method of site suitability analysis utilized in
this study is reviewed.
2.1 Review of Relevant Studies
The study is situated firmly in the community of research domain around agroforestry,
agriculture, and geodesign. This scenario-based methodology was intended to provide a holistic
approach to ensure the full range of support information is provided for sustainable development.
A large part of this geodesign process and perspective has been pioneered by Richard Neutra and
Ian Mcharg. The scenario-based study is similar to the geodesign approach. Design and planning
that takes into consideration both environmental and social issues help ensure that our resources
are used appropriately and responsibly, to help us move toward a better future for all
(Dangermond, 2010). However, the study is dissimilar with its analysis of two scenarios. The
criteria within each scenario was determined through empirical means rather than the classic
geodesign community-participatory process and iteration.
Most of the research conducted/available around site suitability analysis of agroforestry and
agriculture does not include a geodesign or scenario-based approach. These studies are focused
on site suitability analysis and the use of remote sensing. The studies within this research domain
predominately use biophysical factors or criteria given their lack of variability. Biophysical
factors tend to remain stable, unlike socioeconomic factors that are affected by social, economic,
and political settings (Vlek et al., 2004). This community of research around the agroforestry or
  11

agriculture domain rarely accounts for anthropogenic phenomena outside of land use land cover
datasets.
2.2 Site Selection for Scenario-Based Land Evaluation
The partnership between local organizations and Galapaguenos agroforestry managers
within the agricultural zone has helped create successful local biodiversity-based businesses that
have conserved natural resources and introduced alternative crops into the global and regional
consumer food markets. Some of these alternative crops include the Wild Galapagos Tomato,
Galapagos coffee, and exotic fruits. The economic value of crops grown in the Galapagos has the
potential to grow along with the appreciation of agroforestry as a science, management system,
and biodiversity conservation measure. The Galapagos National Park and many local
organizations who partner with Galapagos agrarian communities have come to realize that
scaling up conservation of agroforestry with the development of tourism in a sustainable manner
is viable, challenging, and will require mapping along with integrative land-use planning. This
makes the non-protected zones (urban coastal zone and agricultural zone) of the Galapagos a
critical area for land evaluation-site suitability analysis to support balanced land allocation
decisions.
The capabilities of GIS are well suited to provide these mapping and land use planning
needs. Most business development organizations based in Ecuador are familiar with GIS and use
it for inventory mapping purposes but lack studies within a community of research to pull from
to conduct more in-depth geoprocessing and modeling. This gap in deep spatial analysis and
modeling research is precisely the reason why an agroforestry and tourism scenario-based site
suitability assessment study is critical and worthwhile.
  12

2.3 Methods for Site Suitability Analysis
This SSA method is distinctive as it ascertains measurable potential for the expansion or
reorganization of land use for agroforestry and/or tourism within the agricultural zone of the
Galapagos Islands.
To investigate and find suitable locations, a user can overlay several layers in a GIS. This
overlaying method for suitability evaluation was first devised and used by Ian McHarg. He is
well known for his seminal work, Design With Nature where he showed how a user could
superimpose a set of transparent layers, one for each criterion, to create an overall suitability map
(McHarg, 1994). This technique is regarded as a precursor of modern GIS overlay (Qiu et al.,
2014).
There are many methods available today for modeling suitability. A common method for
modeling suitability divides locations being measured into two sets: those that are suitable and
those that are not. This method is known as the Boolean overlay and it evaluates whether a
location meets each criterion, on a yes/no basis (Mitchell, 2012). This method works well when
the attributes or spatial boundaries of a criterion are crisp. However, if attributes and boundaries
of criterion are not crisp, which is the case for this study, two common methods that can be used
to rate a location on a scale from more suitable to less suitable. The two methods are fuzzy
overlay and weighted overlay.
A weighted overlay allows users to assign importance to a specific criterion. When a user
assigns importance, a weight is assigned to the layer (Mitchell, 2012). This assignment of
importance can also be understood as a process of setting percentages of influence for each layer.
In the context of a raster cell within GIS, cell values of each layer inputted are multiplied by their
  13

percentage influence and the results are summarized together to create the final output raster.
This non-crisp method was not used in the study because the information that can be used to
assign the weights is not known for each specific context in which it might be applied. The
framework developed in this study also does not function optimally when setting weights in a
way that can be applied universally.
However, the fuzzy overlay suitability analysis method was found to be well suited and was
utilized for this study. The fuzzy overlay method ranks and combines hard to quantify data using
mathematical or logical functions to produce a scale of suitability (Mitchell, 2012). It is a method
that is particularly good for creating a suitability model that attempts to capture the knowledge of
experts in a particular field. Fuzzy logic is built on the concept of fuzzy sets which allow partial
membership within a range of 0 to 1 when representing the extent to which an entity belongs to a
certain class. This method assigns importance to raster cells assessed within the GIS based on a
measurement of belonging or scale of membership which is reclassified as an appropriate scale
of suitability.
This suitability method and scale illustrate the potential for targeted land allocation of
agroforestry and tourism land use across the agricultural zones of the Galapagos Archipelago
(Isla San Cristobal, Santa Cruz, and Isabela). The scale used with suitability analysis is well
established and developed by the United Nations, Food and Agricultural Organization (FAO) for
land evaluation purposes. Land evaluation is a process for matching the characteristics of land
resources for certain uses using a scientifically standardized technique (Ritung et al., 2007). The
land suitability classification, using the guidelines of FAO (1976) is divided into Order, Class,
Sub Class, and Unit. Order is the global land suitability group. Land suitability Order is divided
into S (Suitable) and N (Not Suitable). Class is the land suitability group within the Order level.
  14

Based on the level of detail of the data available, land suitability classification is divided into
different orders and in this case its divided with the S order common to semi detailed maps (scale
1:25.000-1:50.000) which is comprised of Highly Suitable (S1), Moderately Suitable (S2),
Marginally Suitable (S3) and Not Suitable (N) (Ritung et al. 2007). The number of classes is
determined based on the number of characteristics/variables being evaluated and the level of
database detail.
This study utilizes semi-detailed maps and an extensive database of variables for both
scenario 1 and 2. The SSA is measured and qualitatively standardized using four classes of the S
order as listed (table.1). The four classes of S order are converted from a fuzzy overlay model
output measurement ranging from a scale of 0 to 1 which is discussed further in chapter three
and into four distinct categories or suitability classes.
Table 1. The site suitability classification system used for land evaluation (FAO guidelines
(1976)) (Ritung et al., 2007).

Suitability Class Description
S1
Highly Suitable: Land having no significant limitation or only
have minor limitations to sustain a given land utilization type
without significant reduction in productivity or benefits and will
not require major inputs above acceptable level.
S2
Moderately Suitable: Land having limitations which in aggregate
are moderately severe for sustained application of the given land
utilization type; the limitations will reduce productivity or
benefits and increase required inputs to the extent that the overall
advantage to be gained from the use, although still attractive, will
be appreciable compared to that expected from Class S1 land.
  15

S3
Marginally Suitable: Land having limitations which in aggregate
are severe for sustained application of the given land utilization
type and will so reduce productivity or benefits, or increase
required inputs, that any expenditure will only be marginally
justified.

N Not Suitable as the range of inputs required is unjustifiable.

  16

CHAPTER 3: A Scenario-Based Land Evaluation Framework for Biodiversity
Conservation in the Galapagos Archipelago

The chapter describes the study area and its geographic properties. Then it discusses the
attributes of data collected, prepared, and utilized. It continues with a discussion of fuzzy overlay
processes, the definition of the land evaluation scenarios, ground-truthing as validation, the
suitability criteria, and the geoprocessing workflow used. Finally, the major and minor model
criteria within the framework are reviewed.

3.1 Study Area
The total geographic extent of the Galapagos Islands is over 799,313 ha, of which only
25,059 ha (61,922 acres) are designated for agriculture and referred to as the agricultural zone
(fig. 1). All four inhabited islands of the Galapagos, Santa Cruz, San Cristobal, Isabela, and
Floreana, have a zone in the humid highlands that has been designated for agricultural use
(Colloredo-Mansfeld et al., 2020). These agricultural zones face the south or windward side of
the islands, which receive high levels of precipitation during the warm season (January–May)
and remain enveloped in clouds during the cool season (June–December) (Colloredo-Mansfeld et
al., 2020).
The Galapagos National Park (GNP) charged with managing and protecting the Islands
ecology, is in control of 97 percent of the terrestrial area. While the local community lives in the
non-protected 3 percent referred to as the agricultural zone which is fully surrounded by the
protected region. Inhabitants of the agricultural areas of the Galapagos devote their land to three
general activities: cattle ranching (bovine, poultry, pork), crop production (permanent and annual
crops), and tourism activities (Colloredo-Mansfeld et al., 2020).
  17

The Galapagos Islands was selected as the area of interest for three main reasons: the
autonomous region is experiencing rapid tourism development that is affecting the quantity and
quality of essential natural resources, the local community is interested in agroforestry methods
of biodiversity conservation, and open-source geospatial data is available. An area of interest
with open-source geographic data will help ensure that the study can be replicated in a variety of
other locations of similar contexts.
3.2 Data
The suitability data collected, prepared, and analyzed in this study was categorized as either
critical or background data. These critical datasets consist of digital elevation, soils, aridity,
potential evapotranspiration, agricultural zone polygon delineation, average annual temperature,
average annual precipitation, and land use land cover data. While the essential background
datasets consist of open street map roads, open street map derived market center points, open
street map derived coastal port route points, and satellite imagery (table. 2). All datasets were
standardized to a raster cell resolution of 30 arc seconds. This standardization of spatial dataset
resolution was executed through the resampling operation within the ArcGIS Pro platform. The
standardization of these critical and background datasets ensured the highest possible accuracy of
criteria or environmental phenomena representation while allowing all datasets to be compatible
for combined analysis.

Table 2. Summary of Data
Data Author Temporality Data Type Resolution
Extent of
Coverage
Content
Agricultural Zone
Polygon Data
Petros Maskal
(created in
ArcGIS Pro)
2020 Vector N/A Regional
This is a boundary feature layer
that delineates the agricultural
zone.
Digital Elevation
Data
CGIAR-CSI
GeoPortal
developers
2004-2020 Raster 90 m Global
This is a digital elevation model
(DEM) shapefile originally
obtained in TIFF format with
aspatial elevation measurements.
A total of three SRTM DEM
frames were collected to capture
the full Galapagos Island surface
area. The majority of slope and
elevation datasets are developed
from this DEM data source.
Open Street Map
Data
Open Street
Map
community
2020
Vector (raster
transformation)
30 arc sec Global
This is a road polyline shapefile
for the Galapagos region.
ISRIC-FAO Soils
Data
International
Soil Reference
and
Information
Centre group
2017
Vector (raster
transformation)
30 arc sec
South
American
Continent
This is a vector shapefile of South
America and the surrounding
Islands that contains soil texture,
symbology and descriptive aspatial
data.
1 8

CGIAR-CSI
Potential Aridity
Data
CGIAR CSI
Consortium
for Spatial
Information
1982-2012 Raster 30 arc sec Global
This is an aridity (combined
factors of annual rainfall and
temperature) shapefile originally in
TIFF format with aspatial aridity
metrics. The global Landsat based
frame was collected to capture the
full Galapagos Island surface area.
CGIAR-CSI
Potential
Evapotranspiration
(PET) Data
CGIAR CSI
Consortium
for Spatial
Information
1982-2012 Raster 30 arc sec Global
This is a PET (combined factors of
annual evaporation, transpiration
and temperature) shapefile
originally obtained in TIFF format
with aspatial evapotranspiration
metrics. The global Landsat based
frame was collected to capture the
full Galapagos Island surface area.
Coastal Port Routes
Petros Maskal
(created in
OSM)
2020
Vector (raster
transformation)
30 arc sec Global
This is a point feature layer that
marks all the major coastal ports
and docks of the Galapagos
Archipelago.
Road Network Data
Petros Maskal
(created in
OSM)
2020
Vector (raster
transformation)
30 arc sec Global
This is a polyline feature layer that
marks all the primary and
secondary roads of the Galapagos
Archipelago.
Land Use Land
Cover (LULC) Data
European
Space Agency
Group
2018 Raster 30 arc sec
South
America,
West
Africa
This is a LULC shapefile classified
using the USGS standard system.
The dataset contains aspatial
LULC measurements.
1 9

Average Annual
Temperature Data
WorldClim
Group
2018 Raster 30 arc sec Global
This is an average annual
temperature shapefile originally
obtained in TIFF format with
aspatial temperature metrics. The
global WorldClim based frame
was collected to capture the full
Galapagos Island surface area.
Average Annual
Precipitation Data
WorldClim
Group
2018 Raster 30 arc sec Global
This is an average annual
precipitation shapefile originally
obtained in TIFF format with
aspatial precipitation metrics. A
global WorldClim based frame
collected to capture the full
Galapagos Island surface area.
2 0
  21

3.3 Methodology
Many methods of site suitability analysis exist and several of these methods have been
considered and described in chapter two. This section describes the fuzzy overlay method in
more detail which is most appropriate for this study, the geoprocessing workflow of the model
framework, details of modeling criteria, and a summary of data attributes.
3.3.1 Fuzzy Overlay Process
This GIS platform and the fuzzy membership functions or fuzzy logic methodology enabled
multicriteria analyses which includes physical and cultural geography factors within the study
area like slope, aspect, elevation, and land cover among other properties.
This study also factored in the established local and community-based research around
economic growth trends and land use parameters within the agricultural zones of Isla San
Cristobal, Santa Cruz, and Isabela. This GIS fuzzy suitability overlay modeling was executed
using a fuzzification framework that assigns importance to layers/model factors through fuzzy
membership sets using sub-model criteria. These major categories or factors of landscape and
geographical characteristics critical to Galapagos tourism development or scenario 2 and
agroforestry production or scenario 1 are in the form of curated datasets organized within the
framework (fig. 3). These spatial datasets/layers within each category were assigned scores and
ranked through the reclassification process. The fuzzy logic procedure of assigning membership
was then executed on these reclassified values using criteria parameters validated by research
and relevant to the study area. The model computed scores and membership values for each
raster cell in the study area resulting in a spatial output range of S1, S2, S3, and N for
agroforestry production or scenario 1 and tourism development or scenario 2.
  22

Figure. 3 The site suitability analysis workflow for both scenario 1 and scenario 2.

This site suitability method of fuzzification was more appropriate for the study when
compared to Boolean overlay or graduated screening methods. The fuzzification method allows
for land evaluation where hard boundaries may not exist, and predictive accuracy is required.
These fuzzy logic functions also provide a clear advantage over Boolean overlays and graduated
screening in their ability to allow for customizable sub modeling of nuanced or continuous
variables like seasonal rainfall fluctuations.
The fuzzy models achieve better predictive accuracies than their classic counterparts. By
incorporating fuzzy suitability membership of environmental factors in the modeling process,
these fuzzy models also produce more informative fuzzy suitability maps. (Qui et al., 2014). The
result of more informative suitability maps increases the effectiveness of land evaluation, land
use decision making, and land allocation for scenario 1 and 2.

  23

3.3.2 Land Evaluation Scenarios
Both the agroforestry and tourism land use were evaluated for their suitability within the
Galapagos agricultural zone. The models for scenario 1 and scenario 2 used a select set of
elevation, aspect, slope, aridity, drainage, evapotranspiration, temperature, land use land cover,
and built environment layers.
Scenario 1: Agroforestry Production
The scenario consists of annual and perennial crop and timber production with the most
common tree stands of Scalesia pedunculata, Scalesia cordata also referred to as the Scalesia
forest, and Galapaguenos in this region commonly produce cocoa, coffee, or livestock. Most
Galapaguenos under this land use generate a portion of their income by selling harvested crops
and timber at the local market or by providing forms of agritourism to visitors. This land use has
been known to suppress invasive weed growth and as a result, can conserve the surrounding
natural resource capital.
Scenario 2: Tourism Development
The scenario consists of impermeable surfaces of bare soil, concrete structures, village
centers, vertical buildings that house, and entertain tourists. The Galapaguenos in this scenario of
land use rely on public utilities and generate a portion of their income from dining,
accommodations, or recreational hospitality.
3.3.3 Ground Truthing
A collection of location points with general land ownership boundaries, site photos, and
observations were collected within the agricultural zone across all three Islands of Santa Cruz,
  24

San Cristobal, and Isabela to verify tourism and agroforestry land use patterns in combination
with the data acquired.
The occurrence of agroforestry production was successfully verified across four
locations and tourism development at one location within the agricultural zone of Santa Cruz
Island (fig. 4). These observation site photos in figure 4 of agroforestry production starting from
the top are depicting a silvopasture agroforestry system for cattle grazing, second was a mixed
vegetable polyculture agroforestry system, third was a shaded coffee agroforestry system, and
fourth the intercropping agroforestry system of both perennials like banana plantings, and
annuals like pineapple plantings. The observation site photo on the right of figure 4 is depicting
the front entrance of a Galapagos eco-lodge for tourists.

Figure 4. The site map of four agroforestry locations and one tourism location recorded within
the agricultural zone of Santa Cruz Island, near and within Bella Vista (sites observed in the
month of November 2020).
  25

The occurrence of agroforestry production was successfully verified across three locations
and tourism development at one location within the agricultural zone of Isabela Island (fig. 5).
These observation site photos in figure 5 of agroforestry production from the top are depicting an
orchard agroforestry system, second was an intercropping agroforestry system of papaya with
annual vegetables, and third was an agroforestry system of shaded coffee with banana plantings.
The site observation photo for tourism development in figure 5 is depicting the front entrance of
an eco-lodge.

Figure 5. The site map of three agroforestry locations and one tourism location recorded within
the agricultural zone of Isabela Island, near and within the rural homesteads (sites observed in
the month of November 2020)

The occurrence of agroforestry production was successfully verified across three locations
and tourism development at three locations within the agricultural zone of San Cristobal Island
(fig. 6). These site observation photos in figure 6 for agroforestry production from the top are
depicting a silvopasture agroforestry system of dense tree plantings, second was a polyculture of
  26

annual vegetables with trees, and third was an intercropping agroforestry system of annual corn
and perennial fruit trees. The site observation photos in figure 6 of tourism development from the
top depicts a seafood restaurant catering to tourists, second was a local ice cream business
catering to tourists, and third was an open-air pizza restaurant.

Figure 6. The site map of three agroforestry locations and three tourism locations recorded
within the agricultural zone of San Cristobal Island, near and within the rural homesteads (sites
observed in the month of November 2020).

  27

3.3.4 Suitability Criteria for the General Framework
This section covers the suitability criteria and data summary (table. 5) identified for use in
the framework. The categories of criteria were ranked as either being major or minor model
parameters based on supporting research described in the following chapter sections for both
scenarios 1 and 2. A total of ten criteria were analyzed in scenario 1 contained within six
categories of hydrology, topography, built environment, temperature, soils, and solar energy in
relation to the aspect (table. 3). While scenario 2 includes an analysis of five criteria contained
within two categories of temperature and built environment (table. 4).

Table 3. Scenario 1 Site Suitability Analysis Criteria
Category Criteria ID Criteria Summary Statement for Selection of Criteria
1. Avoid major
hydrological
constraints on
agroforestry
production
1a
Should not be established
within low potential
evapotranspiration zones
The potential evapotranspiration provides a clear indication of
healthy active plant communities through the measurement of
gas exchange.
1b
Should not be established on
terrain that receives a low
quantity of precipitation
Annual precipitation is a major factor that directly affects plant
survival and overall health
1c
Should not be established in
areas of high aridity
The combination of chronic high temperatures with low rainfall
produces an arid climate which is a climate not conducive for
large plant community growth.
2. Select regions
where the
topography is
adequate for
agroforestry
production
2a
Should be established within
the optimum altitude where
the humid and transition zone
exists
The elevation and altitude of terrain are a major factor in plant
community assemblage (ecozones) in addition to the quantity of
ecological interaction.
2b
Should not be established on
steep slopes that are difficult
to access, plant or harvest
Terrain with a certain percent slope or greater can be difficult to
access and operate for crop production without causing soil
degradation and a resulting loss of yield.
2 8

3. Avoid regions
with a built
environment where
development exists
3a
Should not be established
within the urban land use
The establishment of plant communities in the urban
environment is avoided due to growth limitations from
impermeable surfaces and the shadow effect of buildings.
3b
Should not be established
within primary and
secondary roads
The establishment of plant communities within the road
network is avoided due to growth limitations from impermeable
surfaces.
4. Avoid regions
with higher or lower
than average
temperature
4a
Should not be established in
regions with high or low
annual temperature
High or low long-lasting temperature (more than a week) is
known to damage and weaken most plants.
5. Select a region
where soils provide
optimal drainage
5a
Should be established in soils
with an optimal drainage
class range
A well-drained soil with moderate porosity and permeability
allows for adequate plant community rootzone growth.
6. Avoid terrain with
low solar energy
related to the aspect
6a
Should not be established on
aspects with low annual
sunlight
The specific direction of terrain in relation to the sun has an
effect on photosynthetic energy potential for most plant
communities.
2 9

Table 4. Scenario 2 Site Suitability Analysis Criteria
Category Criteria ID Criteria Summary Statement for Selection of Criteria
1. Select regions
near the existing
built environment
and Infrastructure
3c
Should be
developed near
primary and
secondary roads
The placement of tourism facilities and structures near major roads ensures
public visibility and provides ease of mobility for tourists exploring the
Island.
3d
Should be
developed in
proximity to a
market center
The quick access to restaurants, cafes, bars and the potential site seeing
benefits in market centers can increase tourism traffic.
3e
Should be
developed near or
within the existing
urban land use
The development of tourism facilities and structures near existing urban
development increases the probability of having access to existing public
utilities like water or gas lines and can make other building requirements
more affordable. This also has the potential to decrease the amount of urban
land use expansion.
3f
Should be
developed in
proximity to a
coastal port
A tourism development in close proximity to coastal ports or docks can cut
down on travel time for tourists traveling by boat which is one of the main
modes of tourist access to the Islands.
2. Avoid regions
that contain higher
or lower than
average
temperatures that
may be
uncomfortable for
recreation
4b
Avoid regions that
contain higher or
lower than
average
temperature for
the majority of the
year
A tourism development location that holds higher or lower than average
outside temperature year-round can be undesirable to explore, physically
uncomfortable, and cut down on tourist traffic.
3 0

Table 5. Data Summary and Attributes of
Analysis Criteria
Criteria
ID
Map Layer How Data were Created Source Resolution Extent of Coverage Accessibility
1a
Potential
Evapotranspiration
Data developed by
NASA using remote
sensing classification
tools computed from
primary data collected
by orbiting satellites.
Government/NGO
(CGIAR-CSI)
30 arc sec Global Public
1b
Average Annual
Precipitation
Data developed by
NASA using remote
sensing classification
tools computed from
primary data collected
by orbiting satellites.
Government/NGO
(NASA,
WorldCim)
30 arc sec Global Public
1c Potential Aridity
Data developed by
NASA using remote
sensing classification
tools computed from
primary data collected
by orbiting satellites.
Government/NGO
(CGIAR-CSI)
30 arc sec Global Public
2a Terrain Elevation
Data developed using
geo-processing tools
computed from original
dataset.
SRTM-CGIAR 90 m Global Public
3 1

2b Slope of Terrain
Data developed using
geo-processing tools
computed from original
dataset.
SRTM-CGIAR 90 m Global Public
3a, 3e
Land Use Land
Cover
Data developed by the
European space agency
using remote sensing
classification tools
computed from primary
data collected by
orbiting satellites.
European Space
Agency
30 arc sec
South America,
West Africa,
Western Siberia
Organization
3b, 3c Road Network
Data digitized from
secondary data source.
OSM
Vector (raster
transformation
to 30 arc sec)
Global Public
3d Market Center
Data digitized from
secondary data source.
OSM
Vector (raster
transformation
to 30 arc sec)
Global Public
3f Coastal Port Routes
Data digitized from
secondary data source.
OSM
Vector (raster
transformation
to 30 arc sec)
Global Public
4a, 4b
Average Annual
Temperature
Data developed by
NASA using remote
sensing classification
tools computed from
primary data collected
by orbiting satellites.
Government/NGO
(NASA,
WorldCim)
30 arc sec Global Public
5a Soil Drainage
Data digitized from
original dataset.
ISRIC
Vector (raster
transformation
to 30 arc sec)
South American
Continent
Public
3 2

6a Aspect
Data developed using
geo-processing tools
computed from the
original dataset.
SRTM-CGIAR 90 m Global Public
33
  34

3.4 Major Model Criteria and the Implied Constraints
The next two sections describe the major and minor suitability criteria identified for use in
the framework outlined in the previous section. These major criteria for scenario 1 are average
annual precipitation, potential evapotranspiration, potential aridity, terrain elevation, and slope of
the terrain. While the major criteria for scenario 2 are market centers, road networks, land use
land cover, and coastal port routes. These major and minor criteria were differentiated from a
modeling perspective through the GIS fuzzy membership geoprocessing tool of Hedges also
known as the concentration or dilation component of fuzzy membership functions. All the major
criteria were assigned a hedge termed Somewhat or known as dilation which is the square root of
the fuzzification membership function. While all minor criteria were assigned a hedge termed
Very or known as concentration which is the fuzzy membership function squared.
Average Annual Precipitation
The Galapagos Islands agricultural zone is characterized as a humid highland zone with two
seasonal climates which is atypical for the equatorial zone but makes it suitable for year-round
agroforestry. This two-season cycle brings with it a limiting factor of irregular rainfall. The
driver of regional and global rainfall-precipitation across the archipelago is a mix of oceanic
currents, trade winds from the southeast, and the Inter-Tropical Convergence Zone (ITCZ)
movement. This intra-annual ITCZ migration gives rise to the two seasons which characterize
the Galapagos climate: a hot season and a cool season (Trueman et al., 2010). It is in this warm
(hot) season that the agricultural zone receives most of its annual precipitation across all three
Islands of Isabela, Santa Cruz, and San Cristóbal. Average annual rainfall ranges from 500 mm
(20 inches) on the coast to 1500-2000 mm (59-79 inches) in the highlands (above 500 m a.s.l) on
the southern windward side (Taboada et al., 2016). This annual precipitation was visible through
  35

local observation, aerial imagery, and measurements of geologic weathering, soil formation, and
bioclimatic belts of distinct vegetation. This agroforestry suitability of average annual
precipitation within the study area based on this geologic weathering, soils formation, and
bioclimatic indicators was highest within the 22 to 28 inch zone (fig. 7). The agroforestry
suitability within the average annual precipitation criteria decreases with decreasing precipitation
geographically.

Figure 7. A visualization of average annual precipitation criteria for the year 2018 within the
Galapagos Island study area.

This study classified the variation in average annual precipitation into four categories
starting with the highly suitable region utilizing the land evaluation suitability system covered in
chapter two. The average annual precipitation criteria were measured in inches of rainfall-
precipitation in the following order of 22-28 inches, 15-21 inches, 8-14 inches, and 7 inches or
less.

  36

Terrain Elevation
This two-season cycle of high and low precipitation or bioclimatic phenomena has had a
major bio-physical effect on the landscape in the form of clear vegetation zones across the
islands. However, the bioclimatic phenomena would not be possible without the orographic
effect of elevation in terrain which pushes this moisture in the form of clouds higher up into the
atmosphere that eventually allows for cooling in the form of rain. These four general vegetation
zones catalyzed by elevation in terrain or increasing altitude are the semi-arid coast, low shrub-
fern, and grass transition belt the humid south-facing slopes, and lastly, the top caldera locally
referred to as the brown zone section of volcanoes. This altitudinal zonation and rising moisture
followed by rain have also contributed to intensive volcanic rock weathering followed by soil
formation. At the same time, the degree of weathering is related to the bioclimatic zones
described by Stoops (2013 a and b): soils with the lowest degree of weathering, PAy, are located
in the arid coastal zone; in the transition zone (ST), BV y OC soils located at 140-240 m
a.s.l.(459-787 feet) show slightly higher values; soils from CMt, Sr, and CrM develop in the
Scalesia zone (SZ) at altitudes between 240-400 m a.s.l (787-1312 feet).; while soils from CMt,
Sr, and CrM with the largest degree of weathering appear in the brown zone, at elevations higher
than 400 m a.s.l (1312 feet) (Taboada et al., 2016). It is this climatic cycle of annual moisture
movement and geologic weathering catalyzed by elevation change that makes agroforestry
production possible within the Galapagos. The agroforestry suitability based on elevation is also
visibly highest from ariel imagery of vegetation density and on the ground observation within the
440 to 339 m (1444 to 1115 feet) range across the three Galapagos Islands and decreases as you
move lower in elevation or above 499 m (1640 feet) (fig. 8).
  37

Figure 8. A visualization of the terrain elevation criteria within the Galapagos Island study area.

This study classified the variation in elevation into four categories starting with the highly
suitable region utilizing the land evaluation suitability system covered in chapter two. The
elevation criteria were measured in feet (meters) of altitude above sea level in the following
order of 1444-1116 feet, 1115-788 feet, 1445-1772 feet, and 787 feet or less in altitude.
Potential Evapotranspiration
The criteria and geographic phenomenon of evapotranspiration are dependent on many other
geographic phenomena like temperature, elevation, and precipitation but also influences several
geographic phenomena like vegetation composition or soil moisture as it relates to soil
development which is of particular importance in the biodiversity hotspot of the Galapagos
Islands. The vegetation distribution shows direct dependence on altitude, slope, exposition to
trade winds, and level of air moisture (Adelinet et al., 2007). This air moisture is generally a
result of evapotranspiration which is defined as the process of atmospheric removal of water
through evaporation (water moving from the liquid to gaseous state from the earth’s surface) and
transpiration (gaseous water being released from plant leaves or stomata as a function of
metabolism). The assessment and analysis of this atmospheric moisture as a result of
  38

evapotranspiration was vital in understanding the potential of plant productivity or agroforestry
productivity-scenario 1 within the study area. This factor of atmospheric moisture specifically
condensation within the Galapagos Islands directly influences plant growth by providing a
continuous source of water for nutrient and energy exchange. This condensation usually occurs
above 250 m altitude and creates extensive stratus clouds, often down to ground-level, locally
called garua (Hamann 1979, Colinvaux 1984, Nieuwolt 1991). These clouds result in two forms
of precipitation; vertical (rainfall) and occult, the latter consisting of fog that condenses on
vegetation and drips or runs down to the grounds (Trueman et al., 2010). In Galapagos, occult
precipitation can significantly increase the total precipitation amount under dense vegetation
(Trueman et al., 2010).
The study accounted for this atmospheric moisture phenomena through the measurement of
PET defined by the FAO as a measure of the ability of the atmosphere to remove water through
Evapo-Transpiration processes. The PET is classified across a range of 10 classes from 9.8
inches or less of potential atmospheric removal of water per year to 98.4 inches or greater. This
FAO PET classification standard was utilized across the study area to visualize the phenomena
(fig. 9) and determine the site suitability of scenario 1.

  39

Figure 9. A visualization of the potential evapotranspiration criteria measuring potential
atmospheric removal of water per year within the Galapagos Island study area.

The study classified PET into 4 categories with 0.0 inches to 44.3 inches as highly suitable,
44.4 inches to 49.2 inches as moderately suitable, 49.3 inches to 59.0 inches as marginally
suitable, and 59.0 inches or greater as not suitable. However, regions within the 59.0 inches or
greater range do not exist within the study area for this criterion.
Potential Aridity
The criterion of potential aridity is a phenomenon that exists in many regions of the world.
This potential aridity is defined as the ratio of long-term trends in precipitation over long-term
trends of potential evapotranspiration. It is vital to understand and assess the potentiality of this
aridity especially when plant communities or plant production is being considered, given that
plant growth and survival requires conducive atmospheric temperature with adequate water
regimes. The potential effect of aridity on agroforestry production or scenario 1 as it relates to
site suitability was particularly important especially because this scenario was comprised of
long-term perennial plant production along with annual species as discussed in previous sections.
  40

The UN classifies this aridity phenomenon using an index of 5 values with the following
climate classes of hyper-arid at less than 0.03, arid at 0.03 to 0.2, semi-arid at 0.2 to 0.5, dry sub-
humid at 0.5 to 0.65, and humid at greater than 0.65. This study however was largely focused on
the visualization and analysis of arid and semi-arid climate classes within the study area (fig. 10).

Figure 10. A visualization of the potential aridity criteria measuring mean annual precipitation
over mean annual evapotranspiration deriving an index or classes of aridity within the Galapagos
Island study area.

The defining characteristics of the arid and semi-arid climate have been standardized by
Monique Mainguet in the classic droughts and human developments text published in 1999 titled
Aridity. The characteristics of the arid zone as the text defines is its ratio of precipitation over
potential evapotranspiration which equates to an index value of 0.03 to 0.2 and this landscape is
generally comprised of barren areas or those covered by sparse vegetation of perennial and
annual plants. While the semi-arid zone is characterized by a ratio of precipitation over potential
evapotranspiration which equates to an index value of 0.2 to 0.5 and this landscape is generally
covered by steppe open vegetation cover and tropical bush with perennial plants being most
frequent.
  41

The study measured the aridity criteria using this UN-established system of aridity index
values into 4 classes. All humid values at 0.65 or greater were highly suitable, dry sub-humid
values at 0.5 to 0.65 were moderately suitable, semi-arid values at 0.2 to 0.5 were marginally
suitable, and arid values at 0.2 or less were not suitable. However, regions within the 0.65 or
greater and 0.5 to 0.65 range do not exist within the study area for this criterion.
Slope of Terrain
The criteria of slope or ratio of vertical change and also referred to as relief in the landscape
was critical to understanding the site suitability of agroforestry production or scenario 1 within
the study area. This relief is an important factor when it comes to land evaluation overall but
particularly for land use that requires plant growth and management. The relief is related to land
management and erosion hazard and elevation is related to temperature and solar radiation and
thus closely linked to plant requirements (Ritung et al., 2007). This relief-slope factor is linked to
many other geographic phenomena in addition to cultures of management and thus influences the
where of agroforestry production which is a management system of food production.
According to the United States Department of Agriculture (USDA), its most optimal to
produce food and manage production on slopes ranging from 0 percent or flat as ideal, to 30
percent or hilly as the limit. The study utilized a standard FAO land evaluation system of
topography to visualize slope which classified 7 types of relief or percent slope ranging from flat
at less than 3 percent to very steep at greater than 60 percent.
This study however only considered 4 classes given the food production requirements of
scenario 1 (fig. 11).
  42

Figure 11. The criteria and visualization of percent slope of terrain within the Galapagos Island
study area.

The percent slope of 8 or less is classified as highly suitable, 15 or less as moderately suitable, 30
or less as marginally suitable, and 100 percent or less as not suitable.
Market Centers
The market centers of the Galapagos Islands and the accessibility to them was a critical
component in determining the suitability of tourism development or scenario 2 from an
economic perspective. These market centers or hubs of economic activity within the study area
also provide employment and were located in the towns of Bella Vista and Santa Rosa on Santa
Cruz Island or El Progreso on San Cristobal Island. The National Institute of Statistics and
Censuses of Ecuador estimated that 8,772 people were economically active in the archipelago as
of the year 2002 and employed in 18 sectors. These towns or market centers are considered
central hubs for much of these employment sectors. According to their data, the most important
sectors and their percent of island employment were: transport, storage, and communications
(15.3%); vehicle and motorcycle servicing (11.2%); agriculture and ranching (10.3%); public
administration (10.3%), and construction (7.6%) (Epler, 2007). This tourism industry was not
  43

listed as a separate sector but is known to be one of the largest employers in the Galapagos.
Wilen and Stewart (2000) reported that in 1999, 40% of the Galapagos population was employed
within the tourism sector or connected business (Epler, 2007). These market centers function as
centers for employment but also hubs for tourists to exchange essential goods and services (food,
water, toiletries, Wi-Fi, and communication equipment).
This study equated the degree of suitability based on the concept of accessibility to these
market centers which were measured by a range of distance from these market center points or
centroids (fig. 12). The tourism modeling study within the Galapagos authored by Francesco
Pizzitutti and others provided precedence for a hotel, cruise ship, and road network-based
measurement of accessibility at ranges of 0 to 5 miles, between 5 and 6 miles, and 6 miles or
greater. This sort of modeling parameter is effective for measuring accessibility to market centers
given that the market centers or the central plaza phenomena have similar structural properties to
hotels or cruise ship-ports like economic or social centralization.

Figure 12. A visualization of the market center criteria along with its associated euclidean
distance measurements within the Galapagos Island study area.

  44

This study measured and classified the degree of accessibility into three ranges with
locations at the market center point or centroid equating to 0. The zones that are less than 5 miles
out from the point were highly suitable, those within the zone of 5 to 6 miles were moderately
suitable and any zone 6 miles or greater from the point were marginally suitable.
Road Network
The location and arrangement of road networks is a critical factor in determining whether
visitors to the Galapagos Islands continue to have access to pristine sites of world heritage and
convenient accommodations for their stay. Any future Galapagos Island tourism development
plans will require the utilization and visualization of existing road networks-infrastructure (fig.
13).
This study equated the degree of suitability based on the concept of accessibility to roads
which were measured by a range of distance from these roads. The tourism modeling study
within the Galapagos authored by Francesco Pizzitutti and others provided precedence for a
hotel, cruise ship, and road network-based measurement of accessibility at ranges of 0 to 5 miles,
within the zone of 5 to 6 miles, and 6 miles or greater.

  45

Figure 13. A visualization of the primary and secondary road network criteria along with its
associated euclidean distance measurements within the Galapagos Island study area.

This study measured and classified the degree of accessibility into three ranges with
locations on or adjacent (roads were assigned as the centroid) to the road at zero. The zones that
are 5 miles or less out from the roads were highly suitable, those within the zone of 5 to 6 miles
were moderately suitable and any zone 6 miles or greater from the roads were marginally
suitable.
Coastal Port Routes
This coastal port route factor is important to consider, to ensure accessibility to and for fleet
or cruise-based Galapagos Island tourists. The distance and resulting travel time from hotels or
accommodations within the study area to Galapagos Island ports was a criterion critical to
understanding optimal tourism development or scenario 2. A large portion of Galapagos Island
tourists arrive and depart by charter vessels and cruise ships. Most owners cater to the higher
income, predominately foreign tourists. At the other end of the spectrum are vessels oriented
towards budget-minded backpackers and Ecuadorians (Epler, 2007). If less time is spent to and
from locations in this case from the Galapagos Island agricultural zone to coastal ports more time
  46

can be spent touring or exploring locations. Any future Galapagos Island tourism development
plan will require a firm understanding of this distance and time factor in relation to coastal ports.
This study equated the degree of suitability based on the concept of accessibility to the
closest points within the road network to coastal routes. The access to these coastal route points
were measured by the range in euclidean distance from these points in miles (fig. 14). The
tourism modeling study within the Galapagos authored by Francesco Pizzitutti and others
provided precedence for a hotel, cruise ship, and road network-based measurement of
accessibility at ranges of 0 to 5 miles, within the zone of 5 to 6 miles, and 6 miles or greater.

Figure 14. A visualization of points of access to coastal port routes criteria along with its
associated euclidean distance measurements within the Galapagos Island study area.

This study measured and classified the degree of accessibility into three ranges with
locations at the point or centroid equating to 0. The zones that were 5 miles or less out from the
point are highly suitable, those within the zone of 5 to 6 miles were moderately suitable and any
zone 6 miles or greater from the point were marginally suitable.

  47

Land Use Land Cover
The identification and consideration of existing land use land cover (LULC) are vital,
particularly within the Galapagos Island study area. These ecosystems and the unique plant
communities within them are of high importance to maintain biodiversity. The Island’s natural
composition of land cover is generally made up of clear climatic or vegetation zones as
mentioned in previous sections. Whilst these climatic zones have not been mapped, they
correspond to naturally occurring semi-arid and humid vegetation zones as described by Hamann
(1979) and mapped by Huttel (1986) (Trueman et al., 2010). These regions of land cover are
usually referred to as climatic zones rather than vegetation zones partly because it’s a factor of
climate and partly because these zones have been completely altered by anthropogenic change or
land use. To contribute towards arresting this change in natural land cover, future land use like
agroforestry production or scenario 1 must be assessed for its suitability in relation to land use
land cover types-classes that already exist. The three major land use land cover classes that exist
within the study area determined/mapped by the European Space Agency (ESA) are the rainfed
cropland mosaic greater than 50 percent, broadleaf tree cover greater than 15 percent, and open
shrubland (fig. 15).

  48

Figure 15. A visualization of the ESA classified LULC criteria within the Galapagos Island study
area.

This study utilized the ESA’s established LULC system and classified suitability based on
the degree of disturbance. The urban-built environment or bodies of water were most disturbed
or not suitable (classes 0 and 160 to 220), regions with sparse canopy cover or deciduous plant
species were marginally suitable (classes 120 to 153), regions with higher-than-average canopy
cover with broadleaf plant species were moderately suitable (classes 11, 12 and 40 to 110) and
regions currently under crop production were highly suitable (classes 10, 20 and 30) or the least
disturbed for scenario 1 land use.
3.5 Minor Model Criteria
Road Network
The location and arrangement of road networks within the Galapagos Islands determines
largely the accessibility and availability of food resources for its population. This food produced
on the island must be exported off the island or transported to local markets through the conduits
of primary and secondary road infrastructure networks. This road network also plays an essential
role in providing a conduit for transporting imported processed food to the island which
  49

composes the bulk of food resources consumed by Galapagenous or tourists (fig. 13).
Additionally, transport and communications are critical issues, as a failure in either one may
cause uncertainty in the timely supply of products (Sampedro et al., 2018). These products
include scenario 1 of local agroforestry harvested food supply for the community and visitors
alike.
The study provides a system for measuring this degree of accessibility/availability to food
resources by applying the distance and time factor utilized by the author Francesco Pizzitutti and
others in a Galapagos Island tourism modeling study. This system of measurement classifies the
degree of accessibility into three ranges with locations on or adjacent (roads are assigned as the
centroid) to the road at zero. The zones that are less than 5 miles out from the roads are highly
suitable, those within the zone of 5 to 6 miles are moderately suitable and any zone 6 miles or
greater from the roads are marginally suitable.
Soil Drainage
The soil drainage or more specifically the soils hydrodynamic properties is an important
factor that determines what plant diversity (plant root access to water and nutrients) and
assemblages can exist in a particular site and this diversity is necessary for agroforestry
production in the Galapagos Islands as described in previous sections. Some of the key
determining characteristics within the soil drainage factor are soil texture- degree of permeability
(measured as the rate of hydraulic conductivity or water conveyance in meters per second) along
with porosity (measured as the void space within a total volume of soil), dependance on the rate
of rainfall and terrain elevation. The physical properties of the soil are in good agreement with
the variations of the rainfall according to the elevation, which appears as the main factor
controlling the soil development (Adelinet et al., 2007). Understanding that the hydrological
  50

cycle is fundamental for water resource management, soil porosity, permeability, mineral
composition, and particle size are important elements within this cycle (Adelinet et al., 2007).
This variation in soil hydrodynamic properties is heterogeneous across the Galapagos Islands
with the greatest contrast between Isla Santa Cruz and Isla San Cristóbal due in part to the
difference in elevation and rainfall as mentioned in previous sections. Two groups of soils were
identified, with a major difference between them. The first group consists of soils located in the
highlands (> 350 m a.s.l.), characterized by low hydraulic conductivity (< 10
-5
m s
-1
) and low
porosity (< 25%). These soils are thick (several meters) and homogenous without coarse
components. Their clay fraction is considerable and dominated by gibbsite. The second group
includes soils located in the low parts of the islands (< 300 m a.s.l.). These soils are characterized
by high hydraulic conductivity (> 10
-3
m s
-1
) and high porosity (> 35%) (Adelinet et al., 2007).
The major soil types are referred to as Umbric Leptosols (LPq) across Isla Santa Cruz, Dystric
Cambisols (CMd) across Isla Isabela, and Eutric Regosols (RGe) across Isla San Cristóbal
following the international soil reference system (fig. 16).

Figure 16. A visualization of the soil drainage class by the dominant soil type within the
Galapagos Island study area.

  51

The study classified this soil drainage variation within LPq, CMd, and RGe using a standard
USDA-Natural Resource Conservation Service system of seven categories ranging from poorly
drained to excessively drained and lastly into four categories of suitability using the suitability
land evaluation system covered in chapter two. These soil drainage categories were associated
with specific percentages and rates of hydraulic conductivity ranges for the given soil type across
each of the three Islands.
Land Use Land Cover
The identification and consideration of existing LULC are important particularly within the
Galapagos Island study area and built environment. This study area has experienced an
increasing level of land cover or ecosystem disturbance due to an increase in tourism, the
population of Galapagos residents, and invasive species arrival. Urban sprawl is creeping into the
highlands (Epler, 2007). Population growth is out of control, the supply of drinking water
heavily exploited, and more and more vehicles are required to meet demand (Epler, 2007). There
is a limit to the built environment or urban expansion on the island due to land area. It’s
advantageous to utilize current utility infrastructure and land use for tourism development-
scenario 2 before expanding the area of land use. Urban development regulations in the
Galapagos assign a limited geographic extension of the municipal territory to the construction of
new urban infrastructure (Pizzitutti et al., 2016). The study accounts for this built environment
phenomena of limited expansion through the assessment and classification of LULC suitability.
This study utilized the ESA’s established LULC system and classified suitability based on
the degree of disturbance within scenario 2. The bare landscape or non-vegetated virgin ground
was highly suitable (classes 200, 201, and 202), regions with a body of water or ice were not
  52

suitable (classes 0, 210 and 220), the urban regions were moderately suitable (class 190) and all
land cover that was vegetated was classified as marginally suitable (class 160 to 180).
Average Annual Temperature
The component of annual temperature cycles within climate has a major effect on plants and
animals within the terrestrial landscape, this includes most of the plant species and humans-
people along with many other mammals.
These daily and monthly fluctuation in atmospheric temperature that generally equates to
predictable yearly average temperatures is an important factor among others that determines the
life cycle of plants and whether people can successfully inhabit a specific environment. This
factor of temperature which is defined as the average annual temperature in this study also
extends its influence on both scenarios of tourism development and agroforestry production.
The bioclimatic changes of plant diversity and quantity in combination with temperature as
you move longitudinally across the globe or within a region maybe the clearest indication of the
temperature effect on plant productivity. The warm humid tropics and semi-tropical zones of the
globe hold more diverse plant growth and provide longer production periods for plant growth
compared to your hot arid or cold tundra regions of the globe. These global and regional
variations in temperature have a direct effect on the productivity of agroforestry systems across
regions of the world and including the Galapagos Island study area (fig. 17).
  53

Figure 17. A visualization of the average annual temperature criteria for the year 2012 within the
Galapagos Island study area.

The effect of temperature on plants influences people in a similar but more nuanced manner.
This factor of temperature has been shown to affect people or the individual through thermal
stress. The condition of thermal stress occurs when the temperature in a location becomes too
extreme for the human body to handle. This condition of thermal stress is particularly important
to consider in a recreational setting like tourism and its influence on visitation. Temperature is
often used as one key variable to model tourist visitation (Hamilton and Tol 2007; Serquet and
Rebetez, 2011) (Becken, 2012). This rate of tourist visitation can have a direct impact on the
economic success of tourism and its development within the Galapagos Island study area.
The study measures this potential rate of tourist visitation in relation to thermal stress and
potential plant productivity through an established index of thermal comfort which is in a similar
range for both people and plants. This index is based on information about thermal comfort or
thermal component, which are not based on human energy balance of individuals (Matzarakis,
2006) (Zaninovic, 2009). These thermal comfort ranges were classified within the study as 72.0
F to 71.1 F as highly suitable, 71.0 F to 69.1 F as moderately suitable, 72.1 F to 73.0 F as
  54

marginally suitable, and below 69.0 F or above 73.0 F as not suitable. However, regions above
73 F do not exist within the study area for this criterion.
Aspect
Our understanding of the slope aspect is essential in determining the suitable and most
advantageous location for agroforestry production or scenario 1. This geographic component of
aspect plays a large part in the development of microclimates and by extension the habitat for
certain plant communities. The slope aspect plays a vital role in determining soil moisture and
solar radiation and thus soil heat flux between the soil and atmosphere (Bennie et al., 2008;
Chesson et al., 2004). Generally, the slope aspect alters radiation and soil evaporation, and thus
influences plant composition (Clifford et al., 2013; Deak et al., 2017). These variations
considerably affect soil biogeochemical cycles and vegetation patterns (Zhao and Li, 2017)
(Kong et al., 2019). This cascading effect of aspect is particularly important within the
Galapagos Island study area which mostly exists on south or southeast facing slopes within the
humid or transition vegetation zone (fig. 18). The humid zone extends from 200 to 450 m a.s.l.
and was originally covered by the endemic Scalesia tree (Hamann, 1979) (Pryet et al., 2012).

  55

Figure 18. A visualization of the aspect criteria within the Galapagos Island study area.

The study measures aspect by the unit of slope direction ranging from 0 to 360. These
measurements of slope direction are classified using the well-established understanding of
directional variation in solar radiation levels in comparison to the historical vegetation zone
phenomena within the study area. This slope direction between 202.5 and 247.5 was classified as
highly suitable, 157.5 to 202.5 as moderately suitable, 247.5 to 292.5 or 67.5 to 112.5 as
marginally suitable, and all other slope direction measurements as not suitable.

  56

CHAPTER 4: Scenario-Based Suitability Framework Implementation
The chapter begins with an analysis of the fuzzy sub-model implementation results for
scenario 1 followed by a section that discusses the final scenario 1 suitability result. Then, the
fuzzy sub-model implementation results for scenario 2 are analyzed followed by a section
discussing the final scenario 2 suitability result. The chapter ends with an analysis of the
suitability interrelationship of both scenarios 1 and 2.
4.1 Scenario 1: Agroforestry Production
The section of scenario 1 discusses the fuzzy membership sub-model results for all six
factors or categories of individual criteria discussed in chapter 3. This section also covers the
modeling process, source layers, uncertainty in measurements, and the locations that were
determined too definitely be suitable or unsuitable. The scenario 1 section concludes with a
description of the agroforestry production site suitability findings, the resulting visualization, and
an area measurement of site suitability by suitability class.
4.1.1 Agroforestry Production Sub Model Results
The sub-model execution of hydrology factors provided a variety of fuzzy membership
results. These hydrology factors classified as major criteria include average annual precipitation-
1b, potential evapotranspiration-1a, and potential aridity-1c. The sub-model output resulted in a
higher membership value for the average annual precipitation and potential aridity phenomena
within the San Cristobal Island region and a higher membership value for potential
evapotranspiration within the Santa Cruz Island region (fig. 19).
  57

Figure 19. Map of fuzzy membership for criterion 1a, 1b and 1c.

The sub-model of hydrology factors was comprised of spatial datasets from the CIGAR-CSI
and WorldClime organizations and prepared for use within the ArcGIS Pro model builder
platform. These datasets were computed through a series of geoprocessing algorithms starting
with the clip operator followed by the reclassify operator, reclass operator, and lastly the fuzzy
membership operator.
The fuzzy membership linear function was selected for use on all hydrology factors due to
the maximum and minimum characteristics of the phenomena and source layer. All the
hydrology criteria are standardized and visualized to the 30 arc seconds resolution which carries
some inherent uncertainty in its accuracy of representation (appendix. A).
The second sub-model factor of topography provided unique fuzzy membership results.
These topographic factors classified as major criteria include Elevation of Terrain-2a and Slope
of Terrain-2b. The sub-model output resulted in a higher membership value for the elevation of
  58

terrain within the northern portions of Isabela, Santa Cruz, and San Cristobal Island while the
slope of terrain holds higher membership values within the southern portion of Isabela and Santa
Cruz and western portion of San Cristobal (fig. 20).

Figure 20. Map of fuzzy membership for criterion 2a and 2b.
The data and source layers for the topographic factor were acquired from the national
aeronautics and space administration of the United States data portal and prepared for use within
the ArcGIS Pro model builder platform. These datasets were computed through a series of
geoprocessing algorithms starting with the clip operator followed by the reclass operator,
reclassify operator, and lastly the fuzzy membership operator.
The fuzzy membership linear function was selected for use on both topographic factors due
to the maximum and minimum characteristics of the phenomena and source layer. The elevation
and slope of terrain criteria were resampled from its original resolution of 90m to 30 arc seconds
resolution for standardization purposes which can cause some inherent uncertainty in its
accuracy of representation (appendix. B).
The third sub-model built environment factors provided a variety of fuzzy membership
results. These built environment factors classified as major criteria include LULC-3a and the
Road Network-3b. The sub-model output resulted in a higher membership value for the road
  59

network phenomena within the western region of San Cristobal Island while the LULC resulted
in a higher membership value within the southern regions of all three islands (fig. 21).

Figure 21. Map of fuzzy membership for criterion 3a and 3b.
The data and source layers for the built environment factors were acquired from the
European Space Agency (ESA) and Open Street Map (OSM) organization’s data portals and
prepared for use within the ArcGIS Pro model builder platform.
The fuzzy membership linear function was selected for use within the model builder for both
the road network and LULC criteria due to the maximum and minimum characteristics of the
phenomena and source layer. The LULC criteria was remotely sensed and visualized at an
original resolution of 30 arc seconds. This ESA’s LULC classification system and
standardization across a 30 arc seconds cell area was a cause of uncertainty due to the
generalization of phenomena (appendix. C). The road network source layer in comparison carries
some uncertainty due to the inherent positional accuracy of digitized OSM polylines.
The fourth sub-model factor of temperature provided a unique fuzzy membership result. The
temperature factor or average annual temperature-4a was classified as a minor criterion within
the study. The sub-model output resulted in a higher membership value for the criteria 4a
phenomena within the southern region of all three islands (fig. 22).
  60

Figure 22. Map of fuzzy membership for criterion 4a.
The fuzzy membership near function was selected for use within the model builder for the
average annual temperature criteria due to the midpoint and small range of temperature
distribution of the phenomena and source layer. The average annual temperature criteria was
standardized and visualized to the 30 arc seconds resolution which carried some inherent
uncertainty in its accuracy of representation (appendix. D).
The fifth sub-model factor of soils provided a unique fuzzy membership result. The soils
factor or soil drainage-5a is classified as a minor criterion within the study. The sub-model
output resulted in a higher membership value for the criteria 5a phenomena within the Santa
Cruz Island region (fig. 23).

Figure 23. Map of fuzzy membership for criterion 5a.
  61

The fuzzy membership linear function was selected for use within the model builder for the
soil drainage criteria due to the maximum and minimum characteristics of the phenomena and
source layer. The soil drainage criteria is standardized using an established USDA soil drainage
classification system and visualized to the 30 arc seconds resolution which carried some inherent
uncertainty in its accuracy of representation (appendix. E).
The sixth sub-model factor of aspect provided a unique fuzzy membership result. The aspect
factor or aspect criteria-6a is classified as a minor criterion within the study. The sub-model
output resulted in a higher membership value for the criteria 6a phenomena in isolated regions at
the individual cell area level of Santa Cruz and San Cristobal Island (fig. 24).

Figure 24. Map of fuzzy membership for criterion 6a.
The fuzzy membership linear function was selected for use within the model builder for the
aspect criteria due to the maximum and minimum characteristics of the phenomena and source
layer. The aspect criteria were resampled from its original resolution of 90m to 30 arc seconds
resolution for standardization purposes which caused some inherent uncertainty in its accuracy of
representation (appendix. F).

  62

4.1.2 Agroforestry Production Scenario
The fuzzy overlay And operator was used to combine the fuzzy membership layers of
scenario 1. This And operator assigns the minimum values from all the input fuzzy membership
layers to the output cell. The operator identifies the least common denominator of the
membership criteria, producing a more conservative result with smaller overall membership
values. This allows cells with a membership of a specific minimum value of a criterion to be
identified. The fuzzy overlay execution and analysis resulted in clear visualization of site
suitability for agroforestry production classified under the categories highly suitable-S1,
moderately suitable-S2, marginally suitable-S3, and not suitable-N (fig. 25).

Figure 25. The fuzzy overlay suitability result of scenario 1or agroforestry production within the study area-agricultural zone of the
Galapagos Islands.
6 3
  64

This exclusive geoprocessing And function within the fuzzy overlay operator have
disqualified or classified major areas to be not suitable due to the geoprocessing method of
assigning values based on the lowest common denominator for all criteria. The exclusive
function of the fuzzy overlay operator was beneficial because it removed major parts of the
region from consideration allowing for more focused site surveying.
While much of the study area is not suitable, it is not as exclusive as it appears given that
each individual raster cell in the visualization equates to 0.56 by 0.56 miles or approximately
285.4 acres. The site suitability membership class S1 was measured at 4 percent, S2 at 0 percent,
S3 at 16 percent, and N at 80 percent across the entire study area (table. 6).
Table 6. The suitable area measurement and membership class results of scenario 1.
Membership
Range Membership Class Overall Suitability
Approximate
Acres
0.75-1.00 Highly Suitable 4% 2,477.28
0.50-0.749
Moderately
Suitable 0% 0
0.25-0.499
Marginally
Suitable 16% 9,909.12
0.0-0.249 Not Suitable 80% 49,545.60

These highly suitable and marginally suitable sites were in the central region of Santa Cruz
Island and western region of San Cristobal Island equating to a sum of 12,386.4 acres.
4.2 Scenario 2: Tourism Development
The section of scenario 2 discusses the fuzzy membership sub model results for the two
factors or categories of individual criteria discussed in chapter 3. This section also covers the
modeling process, source layers, uncertainty in measurements and the locations that are
determined too definitely be suitable or unsuitable. The scenario 2 section concludes with a
  65

description of the tourism development site suitability findings, the resulting visualization, and
an area measurement of site suitability by suitability class.
4.2.1 Tourism Development Sub Model Results
The sub-model factors of the built environment provided a variety of fuzzy membership
results. These built environment factors classified as major criteria include the Road Network-3c,
Market Center-3d, LULC-3e, and the Coastal Port Routes-3f. The sub-model output resulted in a
higher membership value for the road network phenomena within the western region of San
Cristobal Island while the LULC resulted in a higher membership value within the northern
regions of all three islands (fig. 26). The 3d criterion is similar to 3c as it resulted in a higher
membership value within the western portion of San Cristobal Island while criterion 3f resulted
in a higher membership value across most of Santa Cruz Island, western San Cristobal Island,
and eastern Isabela Island.

Figure 26. Map of fuzzy membership for criterion 3c, 3d, 3e and 3f.
  66

The data and source layers for the built environment factors were acquired from the ESA’s
data portal, OSM data portal, and digitized from the OSM remotely sensed imagery. These
source layers were prepared for use within the ArcGIS Pro model builder platform.
The fuzzy membership linear function was selected for use within the model builder for all
the built environment criteria due to the maximum and minimum characteristics of the
phenomena and source layer. The road network, coastal port routes, and market center source
layer carried some uncertainty due to the inherent positional accuracy of digitized OSM
polylines and central points. The LULC criteria in comparison were remotely sensed and
visualized at an original resolution of 30 arc seconds. This ESA’s LULC classification system
and standardization across a 30 arc seconds cell area was also a source of uncertainty due to the
generalization of phenomena (appendix. G).
The second sub-model factor of temperature provided a unique fuzzy membership result.
The temperature factor or average annual temperature-4b was classified as a minor criterion
within the study. The sub-model output resulted in a higher membership value for the criteria 4b
phenomena within the southern region of all three islands (fig. 27).

Figure 27. Map of fuzzy membership for criterion 4b.
  67

The fuzzy membership near function was selected for use within model builder for the
average annual temperature criteria due to the midpoint and small range of temperature
distribution of the phenomena and source layer. The average annual temperature criteria was
standardized and visualized to the 30 arc seconds resolution which carried some inherent
uncertainty in its accuracy of representation (appendix. H).
4.2.2 Tourism Development Scenario
The fuzzy overlay And operator were used to combine the fuzzy membership layers of
scenario 2. This And operator assigns the minimum values from all the input fuzzy membership
layers to the output cell. The operator identifies the least common denominator of the
membership criteria, producing a more conservative result with smaller overall membership
values. This allows cells with a membership of a specific minimum value of a criterion to be
identified. The fuzzy overlay execution and analysis resulted in clear visualization of site
suitability for tourism development classified under the categories of highly suitable-S1,
moderately suitable-S2, marginally suitable-S3, and not suitable-N (fig. 28).

Figure 28. The fuzzy overlay suitability result of scenario 2 or tourism development within the study area-agricultural zone of the
Galapagos Islands.
6 8
  69

This exclusive geoprocessing And function within the fuzzy overlay operator have
disqualified or classified major areas as not suitable particularly within San Cristobal Island due
to the geoprocessing method of assigning values based on the lowest common denominator for
all criteria. The exclusive function of the fuzzy overlay operator was beneficial because it
removed major parts of the region from consideration but also determined major areas to be
moderately suitable allowing for more focused site surveying.
While a good portion of the study area was not suitable, it was not as exclusive as it appears
given that each individual raster cell in the visualization equates to 0.56 by 0.56 miles or
approximately 285.4 acres. The site suitability membership class S1 was measured at 5 percent,
S2 at 50 percent, S3 at 3 percent, and N at 42 percent across the entire study area (table. 7).
Table 7. The suitable area measurement and membership class results of scenario 2.
Membership
Range Membership Class Overall Suitability
Approximate
Acres
0.75-1.00 Highly Suitable 5% 3,096.60
0.50-0.749
Moderately
Suitable 50% 30,966.00
0.25-0.499
Marginally
Suitable 3% 1,857.96
0.0-0.249 Not Suitable 42% 26,011.44

These highly suitable sites were located predominately in the center of San Cristobal Island.
While the moderately suitable and marginally suitable sites were located across the majority of
Santa Cruz and Isabela Island. All the suitable sites of S1, S2 and S3 combined equate to
35,920.6 acres.

  70

4.3 Interrelationship of Scenarios
The fuzzy membership classes for both scenarios 1 and 2 were defuzzified to produce
quantifiable crisp binary results of suitable and unsuitable (not suitable) sites. Both scenarios
were assessed within the study area to determine the percentage of suitable and unsuitable sites.
The points of spatial overlap for both scenarios were also assessed and measured in acres of both
suitable or both unsuitable sites (table. 8).
Table 8. The defuzzification percentage of suitable area verse unsuitable area for both scenario 1
and 2 compared with area percentages that contain overlap of both scenarios.
Suitable Unsuitable Both Suitable Both Unsuitable
Scenario One 20% 80%
21% 42%
Scenario Two 58% 42%
Approximate Acres: 12,843.00 25,971.00

This assessment of spatial overlap of suitable scenarios was determined to be 21 percent of
the study area or 12,843.0 acres and can be beneficial for more focused site surveying and on the
ground investigation of scenario interrelationships or compatibility. The defuzzification of
scenarios also provided a clear quantitative measure of regions or sites that can be removed or
disqualified from immediate surveying. This unsuitable area for both scenarios or region of
disqualification equates to 42 percent of the study area or 25,971.0 acres.

  71

CHAPTER 5: Summary and Conclusion
This study developed a general framework to identify the suitable sites for where risks of
agroforestry production and tourism development are minimized, and benefits are maximized
within the agricultural zone of the Galapagos Islands. In summary, the research had a threefold
purpose of building upon current research on agroforestry production potential, scenario-based
land use planning in relation to biodiversity conservation, and fuzzy methods of suitability
analysis; creating a suitability framework that could be implemented by governments and NGOs
throughout the humid highland ecosystems while also developing a refined model that applies
the framework to the specific context of the Galapagos Islands, Ecuador. This research included
a review of relevant literature on the risks to biodiversity along with conservation solutions, a
means to evaluate criteria, and developing methods to analyze the criteria in a GIS.
5.1 Assessment of Model
Fuzzy logic was an effective method for this application because it eliminated a false
appearance of certainty in the data. This fuzzy logic allowed for low certainty to be captured.
Fuzzy methods were utilized to represent the fuzziness that was inherent in the data. These
techniques were especially useful for data layers that needed to be resampled to ensure
compatibility for multi-criteria analysis.
5.2 Future Work
While the model implementation showed good performance, it can be further improved.
Field research and further investigation into the regions that scored as not suitable-unsuitable on
the membership scale are needed. This can provide more information about why regions were
  72

deemed unsuitable and their biodiversity significance. Field research can also help to identify
inappropriate criteria in the general framework.
Once a field check is completed, the results of a model, such as the suitability map created in
this scenario-based study, can be used as a decision support tool for biodiversity conservation
planning. The model was able to identify unsuitable areas for scenario 1 and scenario 2 or
regions of development risk to biodiversity based on multiple criteria. These criteria can be
modified and improved as higher resolution data and more types of data become available. The
general framework and the implemented model outlined here are intended to serve as a
foundational tool that local governments and NGO practitioners can continue to improve and
utilize for the valuation of land specific to scenarios or for general planning.
The economic and ecological valuation of land is contextual and can be assessed through a
market-based approach, ecosystem services approach or using simple cost-benefit analysis but
the suitability of locations is a first step that must be understood. This scenario-based framework
and study was intended to provide this multicriteria modeling perspective and first step of land
evaluation for further in-depth land assessment.
The framework was developed remotely and can therefore continue to be improved with
local knowledge and the addition of other scenarios important to the Galapaguenos. It is likely
that as research continues more criteria and scenarios will be added to the framework to account
for the rich and complex planning environment. A user on the local field level with an
understanding of the complexity may also be able to acquire higher resolution data that will
likely improve model performance.

  73

5.3 Applicability of Research
Since many of the risks to biodiversity also exist outside of the Galapagos Islands, the
framework can be used in other regions. The fifteen criteria were developed by researching
universal risks to biodiversity, geographic requirements, and socioeconomic requirements
specific to scenario 1 and scenario 2 within the Galapagos Island study area. Using the general
framework, a user is easily able to select only relevant scenario-based criteria and parameter
requirements as it relates to biodiversity risk and to add additional ones that may be appropriate.
The use of the fuzzy membership function is an important part of the framework because it
allows a user to set membership ratings based on whether or not a scenario-based criterion is
intended to decrease risk to biodiversity or provide socioeconomic benefit. The shape of the
membership function can be selected for each criterion based on knowledge about the
uncertainty. This study utilized a mix of near, gaussian, and predominantly linear functions due
to the type of uncertainty in criterion. The framework included suitability factors that carry a
risk, and/or a benefit to biodiversity therefore the ability for a user to set each membership value
uniquely with a positive or negative slope based on input values or the major minor hierarchy of
criterion using hedges was critical. The choice of the fuzzy membership function was a key
component of the framework during implementation.
When the framework is implemented, it can improve the workflow used by government or
NGOs to determine where a tourism development or agroforestry production enterprise should
be located to ensure the conservation of biodiversity. The framework allows users to disqualify
locations for consideration that have a higher risk to biodiversity or do not provide high
socioeconomic benefits. The framework also demonstrated how a GIS-based tool can be used to
evaluate several criteria with one suitability map.
  74

The scenario-based model and tool does not however capture all the dimensions involved in
deciding where an agroforestry production or tourism development enterprise should be located.
As mentioned earlier, there are several cultural, political, and social factors that influence the
decisions made. The framework can, however, be used to supplement decision-making as a
macro-level tool.
The work in this project lays the foundation for scenario-based assessment in future
development. These general framework measurements used in the model and criteria can
continue to be refined by users as they customize the model for their context. While the model
performed at an adequate level, the study itself offers some suggestions for how a GIS scenario-
based conceptual framework can be used by governments and NGOs to improve land use
planning in regions with limited access to data that ensure the conservation of biodiversity and
meets SDGs in the long-term.

  75

References
Adelinet, M., J. Fortin, N. d’Ozouville, and S. Violette. “The Relationship between
Hydrodynamic Properties and Weathering of Soils Derived from Volcanic Rocks, Galapagos
Islands (Ecuador).” Environmental Geology 56, no. 1 (2007): 45–58.
https://doi.org/10.1007/s00254-007-1138-3.
Armendáriz, Vanessa, Stefano Armenia, and Alberto Atzori. “Systemic Analysis of Food Supply
and Distribution Systems in City-Region Systems—An Examination of FAO’s Policy Guidelines
towards Sustainable Agri-Food Systems.” Agriculture 6, no. 4 (2016): 65.
https://doi.org/10.3390/agriculture6040065.
Ahmad, Firoz, Laxmi Goparaju, and Abdul Qayum. “Agroforestry Suitability Analysis Based
upon Nutrient Availability Mapping: a GIS Based Suitability Mapping.” AIMS Agriculture and
Food 2, no. 2 (2017): 201–20. https://doi.org/10.3934/agrfood.2017.2.201.
Bamberger, Samantha. Ms. Determining the Suitability of Yak-Based Agriculture in Illinois: A
Site Suitability Analysis Using Fuzzy Overlay. Los Angeles, 2017.
Becken, Susanne. “Measuring the Effect of Weather on Tourism.” Journal of Travel Research,
vol. 52, no. 2 (2012): 156–167., doi:10.1177/0047287512461569.
Colloredo-Mansfeld, Mia, Francisco J. Laso, and Javier Arce-Nazario. “Drone-Based
Participatory Mapping: Examining Local Agricultural Knowledge in the Galapagos.” Drones 4,
no. 4 (2020): 62. https://doi.org/10.3390/drones4040062.
Dangermond, Jack. GeoDesign and GIS – Designing our future. ESRI, 2010.
Epler, Bruce. “Tourism, the Economy, Population Growth, and Conservation ...” Charles
Darwin Foundation Report-Tourism, the Economy, Population Growth, and Conservation in
Galapagos, 2007, www.galapagos.org/wp-content/uploads/2012/01/TourismReport1.pdf.
Galderisi, Adriana, Andrea Ceudech, and Massimiliano Pistucci. “A Method for Na-Tech Risk
Assessment as Supporting Tool for Land Use Planning Mitigation Strategies.” Natural Hazards
46, no. 2 (2008): 221–41. https://doi.org/10.1007/s11069-008-9224-8.
Gardener, Mark R., and Christophe Grenier. “Linking Livelihoods and Conservation: Challenges
Facing the Galápagos Islands.” Global Environmental Studies, 2192-6336, no. 1 (2011): 73–85.
https://doi.org/10.1007/978-4-431-53989-6_6.
Gary Bentrup, Tim Leininger. “Agroforestry: mapping the way with GIS.” Journal of Soil and
Water Conservation 57 (2002): 149A-153A.
file:///C:/Users/Petros%20Maskal/Documents/NGO-
%20WWOOFERS%20Project/Agroforestry_Lit/Gary%20Bentrup%20.pdf
  76

González, José A., Carlos Montes, José Rodríguez, and Washington Tapia. “Rethinking the
Galapagos Islands as a Complex Social-Ecological System: Implications for Conservation and
Management.” Ecology and Society 13, no. 2 (2008): 1–26. https://doi.org/10.5751/es-02557-
130213.
Grenier, Christophe. “Galapagos Report 2011-2012 Human Systems.” Island Cultures.
Galapagos Report. Accessed May 17, 2021. https://www.galapagos.org/wp-
content/uploads/2019/09/11.-HUMAN-SYS-poverty.leon-camba-carvajal.pdf.
Guzman, Juan Carlos, and Jose Enrique Poma. “Bioagriculture: An Opportunity For Island Good
Living.” Human Systems. Galapagos Report. Accessed November 17, 2021.
https://www.cabi.org/wp-content/uploads/Guzman-2015-Bioagriculture-Galapagos.pdf.
Hyneman, Jared. Ms. Developing and Implementing a GIS-Based Framework to Identify Optimal
Locations for Clean Water Wells in Sub-Saharan Africa. Los Angeles, 2014.
Ilbery, Brian, Michael Healey, Julie Higginbottom, and David Noon. “Agricultural Adjustment
and Business Diversification by Farm Households." Journal of the Geographical Association,
vol. 81, no. 4 (1996): 301-310.
Kimura, S., and N. Sinha. “Tomato (Solanum Lycopersicum): A Model Fruit-Bearing Crop.”
Cold Spring Harbor Protocols, vol. 2008, no. 12, (2008): 1-10. doi:10.1101/pdb.emo105.
Kong, Weibo, Yufei Yao, Zhongna Zhao, Xin Qin, Hansong Zhu, Xiaorong Wei, Mingan Shao,
Zhao Wang, Kaiqiang Bao, and Min Su. “Effects of Vegetation and Slope Aspect on Soil
Nitrogen Mineralization during the Growing Season in Sloping Lands of the Loess Plateau.”
CATENA 172 (2019): 753–63. https://doi.org/10.1016/j.catena.2018.09.037.
Kumar, Manasi, Burman Erica. “Editorial Critical/Subaltern Perspective on UN MDGs.’’
Journal of Health Management, (2009): 1-14. http://doi.org/ 10.1177/097206340901100101
Laso, Francisco J., Fátima L. Benítez, Gonzalo Rivas-Torres, Carolina Sampedro, and Javier
Arce-Nazario. “Land Cover Classification of Complex Agroecosystems in the Non-Protected
Highlands of the Galapagos Islands.” Remote Sensing, vol. 12, no. 1 (2019): 65.
https://doi.org/10.3390/rs12010065.
Mainguet, Monique. Aridity: Droughts and Human Development. Springer, 1999.
McHarg, Ian L. Design with Nature. New York: J. Wiley, 1994.
MDG Gap Task Force Report, 2015, 3–4. https://doi.org/10.18356/424f9a27-en.
Mitchell, Andy. The ESRI guide to GIS analysis: modeling suitability, movement, and
interaction. Vol. 3. Esri Press, 2012.
  77

Miguel A. Altieri, Clara Ines Nicholls. “Agroecología y resiliencia al cambio climático:
Principios y consideraciones metodológicas.” Agroecologia vol. 8, no.1 (2013): 7-20.
https://revistas.um.es/agroecologia/article/view/182921
Pilehforooshha, Parastoo, Mohammad Karimi, and Mohammad Taleai. “A GIS-Based
Agricultural Land-Use Allocation Model Coupling Increase and Decrease in Land Demand.”
Agricultural Systems, vol. 130 (2014): 116–25. https://doi.org/10.1016/j.agsy.2014.07.001.
Pizzitutti, Francesco, Carlos F. Mena, and Stephen J. Walsh. “Modelling Tourism in the
Galapagos Islands: An Agent-Based Model Approach.” Journal of Artificial Societies and Social
Simulation 17, no. 1 (2014). https://doi.org/10.18564/jasss.2389. .
Pizzitutti, Francesco, Stephen J. Walsh, Ronald R. Rindfuss, Reck Gunter, Diego Quiroga,
Rebecca Tippett, and Carlos F. Mena. “Scenario Planning for Tourism Management: a
Participatory and System Dynamics Model Applied to the Galapagos Islands of Ecuador.”
Journal of Sustainable Tourism, vol. 25, no. 8 (2016): 1117–37.
https://doi.org/10.1080/09669582.2016.1257011.

Porro, Roberto, Robert P. Miller, Marcos R. Tito, Jason A. Donovan, Jorge L. Vivan, Ralph
Trancoso, Rudi F. Van Kanten, Jorge E. Grijalva, Bertha L. Ramirez, and André L. Gonçalves.
“Agroforestry in the Amazon Region: A Pathway for Balancing Conservation and
Development.” Agroforestry - The Future of Global Land Use Advances in Agroforestry, vol. 9
(2012): 391–428. https://doi.org/10.1007/978-94-007-4676-3_20.
Pryet, Alexandre, Christian Domínguez, Pilar Fuente Tomai, Cédric Chaumont, Noémi
d’Ozouville, Marcos Villacís, and Sophie Violette. “Quantification of Cloud Water Interception
along the Windward Slope of Santa Cruz Island, Galapagos (Ecuador).” Agricultural and Forest
Meteorology, vol. 161 (2012): 94–106. https://doi.org/10.1016/j.agrformet.2012.03.018.
Qiu, Fang, Bryan Chastain, Yuhong Zhou, Caiyun Zhang, and Harini Sridharan. “Modeling Land
Suitability/Capability Using Fuzzy Evaluation.” GeoJournal, vol. 79, no. 2 (2013): 167–82.
https://doi.org/10.1007/s10708-013-9503-0.
Ricketts, Taylor H. “Tropical Forest Fragments Enhance Pollinator Activity in Nearby Coffee
Crops.” Conservation Biology, vol. 18, no. 5 (2004): 1262–71. https://doi.org/10.1111/j.1523-
1739.2004.00227.x.
Ritung, Sofyan, Wahyunto, and Fahmuddin Agus dan Hapid Hidayat. “GUIDELINESLAND
SUITABILITY EVALUATION with a Case ...” Land suitability evaluation with a case map of
Aceh Barat district. Accessed November 16, 2020.
http://apps.worldagroforestry.org/downloads/Publications/PDFS/MN15224.pdf.
Roig-Tierno, Norat, Amparo Baviera-Puig, Juan Buitrago-Vera, and Francisco Mas-Verdu. “The
Retail Site Location Decision Process Using GIS and the Analytical Hierarchy Process.” Applied
Geography, vol. 40 (2013): 191–98. https://doi.org/10.1016/j.apgeog.2013.03.005.
  78

Sampedro, Carolina, Francesco Pizzitutti, Diego Quiroga, Stephen J. Walsh, and Carlos F. Mena.
“Food Supply System Dynamics in the Galapagos Islands: Agriculture, Livestock and Imports.”
Renewable Agriculture and Food Systems, vol. 35, no. 3 (2018): 234–48.
https://doi.org/10.1017/s1742170518000534.
Sanchez, P. A. “Science in Agroforestry.” Agroforestry: Science, Policy and Practice Forestry
Sciences, no. 1 (1995): 5–55. https://doi.org/10.1007/978-94-017-0681-0_1.
Scheyvens, Regina, Glenn Banks, and Emma Hughes. “The Private Sector and the SDGs: The
Need to Move Beyond ‘Business as Usual.’” Sustainable Development, vol. 24, no. 6 (2016):
371–82. https://doi.org/10.1002/sd.1623.
Steffan-Dewenter, I., M. Kessler, J. Barkmann, M. M. Bos, D. Buchori, S. Erasmi, H. Faust, et
al. “Tradeoffs between Income, Biodiversity, and Ecosystem Functioning during Tropical
Rainforest Conversion and Agroforestry Intensification.” Proceedings of the National Academy
of Sciences, vol. 104, no. 12 (2007): 4973–4978. https://doi.org/10.1073/pnas.0608409104.
Taboada, Teresa, et al. “Chemical Weathering in the Volcanic Soils of Isla Santa Cruz
(Galápagos Islands, Ecuador).” Geoderma, vol. 261 (2016): 160–168.
doi:10.1016/j.geoderma.2015.07.019.
Trabucco, Antonio, and Robert J. Zomer. “Global Aridity and PET Database.” CGIAR, February
8, 2019. https://cgiarcsi.community/data/global-aridity-and-pet-database/.
Trueman, Mandy, d’Ozouville, Noemi, “CHARACTERIZING THE GALAPAGOS
TERRESTRIAL CLIMATE IN THE FACE OF GLOBAL CLIMATE CHANGE.” Charles
Darwin Foundation, Puerto Ayora, Galapagos, Ecuador Galapagos Research, no. 67 (2010):
26-37. http://aquaticcommons.org/21473/1/GR_67_p26-37.pdf
Verchot, Louis V., Meine Van Noordwijk, Serigne Kandji, Tom Tomich, Chin Ong, Alain
Albrecht, Jens Mackensen, Cynthia Bantilan, K. V. Anupama, and Cheryl Palm. “Climate
Change: Linking Adaptation and Mitigation through Agroforestry.” Mitigation and Adaptation
Strategies for Global Change, vol. 12, no. 5 (2007): 901–18. https://doi.org/10.1007/s11027-
007-9105-6.
Vlek, Paul, Braimoh, Ademola K., Stein, Alfred. “Land Evaluation for Maize Based on Fuzzy
Set and Interpolation.” Environmental Management, vol. 33, no. 2 (2004): 226-238.
https://doi.org/ 10.1007/s00267-003-0171-6
Wahid, Ahmed, Marguerite Madden, Fikry Khalafa, and Ibtehal Fathy. “LAND SUITABILITY
SCENARIOS FOR ARID COASTAL PLAINS USING GIS MODELING: SOUTHWESTERN
SINAI COASTAL PLAIN, EGYPT.” Journal of Urban and Environmental Engineering, vol. 3,
no. 2 (2009): 73–83. https://doi.org/10.4090/juee.2009.v3n2.073083.
79
Wilkinson, S. R., M. A. Naeth, and F. K. Schmiegelow. “Tropical Forest Restoration within
Galapagos National Park: Application of a State-Transition Model.” Ecology and Society, vol.
10, no. 1 (2005). https://doi.org/10.5751/es-01371-100128.
Zachos, Frank E. Biodiversity Hotspots. Springer, 2014.
Zaninovi ć, Ksenija, and Andreas Matzarakis. “The Bioclimatological Leaflet as a Means
Conveying Climatological Information to Tourists and the Tourism Industry.” International
Journal of Biometeorology, vol. 53, no. 4 (2009): 369–374. doi:10.1007/s00484-009-0219-2.

Appendices
Appendix A. Analysis of the scenario one hydrology category of criteria factors.
Criteria Source Layer
Values in source
layer Source of Uncertainty
Definitely
Suitable
Definitely
Unsuitable
Membership
Function
1a-Should not be
established within
low potential
evapotranspiration
zones
CGIAR-CSI,
Potential
Evapotranspiration
The layer consists
of an average
millimeter per
year of potential
evapotranspiration
for years 1982 to
2012 (values
converted to
inches).
The values are
continuous with no
breakpoint and the
layer is visualized in
30 arc sec resolution.
This scale of
visualization and with
no breakpoint or
boundary lines holds
some inherent
uncertainty.
The 20 cells
to the north
on Santa
Cruz Island,
5 cells to the
south on
Isabela and 8
cells in the
central core
of San
Cristobal are
suitable.
The western
region of the
San Cristobal
Island study
area is
definitely
unsuitable.
The linear
function was
used. The
unsuitable
locations
with a
minimum
value of
1499 were
assigned a
membership
value of 0.
Then a
membership
rating of 1
was assigned
to the
maximum
value of
1001.
8 0

1b-Should not be
established on
terrain that
receives a low
quantity of
precipitation
NASA-
WorldClim,
Average Annual
Precipitation
The layer consists
of raster data in
millimeters per
year of average
annual
precipitation for
the year 2018
(values converted
to inches).
There is uncertainty
about how accurately
the layer represents the
phenomena. The layer
is visualized in 30 arc
sec resolution which
can cause most
inherent uncertainty.
The entire
Santa Cruz,
Isabela and
San Cristobal
Island study
area are
definitely
suitable
No locations
of
unsuitability
The linear
function was
used. The
unsuitable
locations
with a
minimum
value of 0
were
assigned a
membership
value of 0.
Then a
membership
rating of 1
was assigned
to the
maximum
value of 700.
1c-Should not be
established in
areas of high
aridity
CGIAR-CSI,
Potential Aridity
The layer contains
aridity index
values comprised
of ten classes:
high aridity of
0.03 as arid to a
low aridity of 1.75
as tundra.
The values are
standardized within the
aridity index and
visualized in 30 arc sec
resolution. This scale
of resolution and data
standardization holds
some inherent
uncertainty in its
accuracy of
representation.
The southern
and eastern
regions of
San Cristobal
Island are
definitely
suitable.
The northern
regions of
Isabela Island
and southern
regions of
Santa Cruz
are definitely
unsuitable.
The linear
function was
used. The
unsuitable
locations
with a
minimum
value of
0.03-class 1
were
assigned a
membership
value of 0.
Then a
8 1

membership
rating of 1
was assigned
to the
maximum
value of
1.75-class
10.
82

Appendix B. Analysis of the scenario one topographic category of criteria factors.
Criteria
Source
Layer
Values in source
layer
Source of
Uncertainty
Suitable
Location(s)
Unsuitable
Location(s) Membership Function
2a-Should be
established
within the
optimum
altitude
where the
humid and
transition
zone exists.
SRTM-
DEM,
layer
The layer contains
landscape altitude
measurements in
meters (values
converted to feet).
The layer was
originally in 90 m
resolution and
resampled to 30
arc sec. This
causes the total
cell area
measurement
accuracy of
altitude to be
uncertain.
The western
region of
Isabela, northern
region of Santa
Cruz and the
northern region
of San Cristóbal
Island are
definitely
suitable.
The area of three
cells in the
southern region
of Santa Cruz
Island is
definitely
unsuitable.
The linear function was
used. The unsuitable
locations with a
minimum value of 459
were assigned a
membership value of 0.
Then a membership
rating of 1 was
assigned to the
maximum value of
1444.
2b-Should
not be
established
on steep
slopes
greater than
30 percent
that are
difficult to
access, plant
or harvest
SRTM-
DEM,
layer
The layer contains
landscape altitude
measurements in
meters (values
converted to feet).
These altitude
values were
converted into
percent slope
using
geoprocessing
tools.
The layer was
originally in 90 m
resolution and
resampled to 30
arc sec. This
causes the total
cell area
measurement
accuracy of
altitude to be
uncertain.
The eastern
region of
Isabela, western
region San
Cristobal and
southern region
of Santa Cruz
Island are
definitely
suitable.
The raster cell
area of a
location in the
western region
of Isabela,
northern region
of Santa Cruz
and southern
region of San
Cristobal Island
are definitely
unsuitable.
The linear function was
used. The unsuitable
locations with a
minimum value of 0
were assigned a
membership value of 0.
Then a membership
rating of 1 was
assigned to the
maximum value of 30.
8 3

Appendix C. Analysis of the scenario one built environment category of criteria factors.
Criteria
Source
Layer
Values in source
layer
Source of
Uncertainty
Suitable
Location(s)
Unsuitable
Location(s)
Membership
Function
3a-Should not
be established
within the urban
land use.
ESA, Land
use land
cover layer
The layer is
comprised of land
use land cover area
measurements that
correspond to a
standard ESA
classification
system of
landscape types
and numerical
codes.
The ESA's remotely
sensed data is
visualized at the 30
arc sec resolution.
This generalization
of the land use land
cover phenomena
across a 30 arc sec
cell area is a cause
of uncertainty.
The
southwestern
region of Santa
Cruz, eastern
region of
Isabela and the
majority of San
Cristobal Island
is definitely
suitable.
The western
region of
Isabela,
northern region
of Santa Cruz
and northern
region of San
Cristobal Island
are definitely
unsuitable.
The linear function
was used. The
unsuitable
locations with a
minimum value of
153 were assigned
a membership
value of 0. Then a
membership rating
of 1 was assigned
to the maximum
value of 10.
3b-Should not
be established
within primary
and secondary
roads
Layer
digitized
from the
OSM
secondary
source
The layer is
comprised of road
network polylines
in association with
euclidean distance
measurements at 5
miles, 6 miles and
greater than 6
miles.
There is uncertainty
in the positional
accuracy of
digitized road
networks and the
exact alignment of
these polylines with
the phenomena of
primary and
secondary roads.
The western
region of San
Cristobal Island
is definitely
suitable.
The eastern
region of San
Cristobal Island
is definitely
unsuitable.
The linear function
was used. The
unsuitable
locations with a
minimum value of
6 were assigned a
membership value
of 0. Then a
membership rating
of 1 was assigned
to the maximum
value of 5 (all
roads-polylines
containing a buffer
of 1312.0 feet of no
data).
8 4

Appendix D. Analysis of the scenario one temperature category of criteria.
Criteria Source Layer
Values in source
layer
Source of
Uncertainty
Suitable
Location(s)
Unsuitable
Location(s)
Membership
Function
4a-Should
not be
established
in regions
with high or
low annual
temperature
WorldClim
Group,
Average
Annual
Temperature
layer
The layer is
comprised of
numerical
temperature
values in degrees
Celsius (values
converted to
Fahrenheit).
The raster layer is
visualized in 30 arc
sec resolution. This
representation of
average annual
temperature at the 30
arc sec scale is a
cause of uncertainty.
The eastern region
of Isabela,
southern region of
Santa Cruz and
western region of
San Cristobal
Island is definitely
suitable.
The western
region of
Isabela Island
is definitely
unsuitable.
The near function
was used. A
membership rating
of 1 was assigned to
the value of 73.4
degrees Fahrenheit
as the midpoint and
membership
decreases on either
side of the midpoint
to an assigned value
of 0 for the value
62.6 and 95.0
degrees Fahrenheit.
8 5

Appendix E. Analysis of the scenario one soil category of criteria.
Criteria Source Layer Values in source layer Source of Uncertainty
Suitable
Location(s)
Unsuitable
Location(s)
Membership
Function
5a- Should be
established in
soils with an
optimal
drainage class
range
ISRIC-FAO,
Soils layer for
the South
American
region
The layer contains
qualitative values of
soil type and area that
are then assigned
standard
measurements-
conductivity of water
in millimeters per hour
(values converted to
inches) of drainage
based on soil type.
The layer value of
dominant soil type is
assigned its
corresponding
hydraulic conductivity
property value and
classified using a
standard drainage class
system of six
categories. This
generalization of
dominant soil type and
its associated drainage
class is a cause of
uncertainty.
The soil
drainage
within Santa
Cruz Island is
definitely
suitable.
The soil
drainage within
Isabela and
San Cristobal
Island is
definitely
unsuitable.
The linear
function
was used.
The
unsuitable
locations
with a
minimum
value of
0.16 were
assigned a
membership
value of 0.
Then a
membership
rating of 1
was
assigned to
the
maximum
value of
4.92.
8 6

Appendix F. Analysis of the scenario one aspect category of criteria.
Criteria Source Layer
Values in
source
layer
Source of
Uncertainty Suitable Location(s)
Unsuitable
Location(s) Membership Function
6a-Should not
be established
on aspects with
low annual
sunlight
SRTM-DEM,
layer
The
values of
slope
direction
from 0 to
360.
The layer was
originally in 90 m
resolution and
resampled to 30
arc sec. This
causes the total cell
area of aspect-
slope direction to
be uncertain.
Three separate cell
locations in the
western portion of
Santa Cruz Island
and two cell
locations within
southern San
Cristobal Island are
definitely suitable.
A large
portion of
Santa Cruz
Island, San
Cristobal
and all of
Isabela are
definitely
unsuitable.
The Gaussian function
was used. A
membership rating of 1
was assigned to the
value of 202.5 as the
midpoint and
membership decreases
on either side of the
midpoint to an assigned
value of 0 for the value
0.0 and 360.0.
8 7

Appendix G. Analysis of the scenario two built environment category of criteria factors.
Criteria
Source
Layer
Values in source
layer
Source of
Uncertainty
Suitable
Location(s)
Unsuitable
Location(s)
Membership
Function
3c-Should be
developed near
primary and
secondary roads
Layer
digitized
from the
OSM
secondary
source
The layer is
comprised of road
network polylines
in association with
euclidean distance
measurements at 5
miles, 6 miles and
greater than 6
miles.
There is uncertainty
in the positional
accuracy of
digitized road
networks and the
exact alignment of
these polylines with
the phenomena of
primary and
secondary roads.
The western
region of San
Cristobal Island
is definitely
suitable.
The eastern
region of San
Cristobal Island
is definitely
unsuitable.
The linear
function was
used. The
unsuitable
locations with a
minimum value
of 6 were
assigned a
membership
value of 0. Then
a membership
rating of 1 was
assigned to the
maximum value
of 5 (all roads-
polylines
containing a
buffer of 1312.0
feet of no data).
8 8

3d-Should be
developed in
proximity to a
market center
Layer
digitized
from the
OSM
secondary
source
The layer is
comprised of
market center
points in
association with
euclidean distance
measurements at 5
miles, 6 miles and
greater than 6
miles.
There is uncertainty
in the positional
accuracy of market
center points
(centers of
commerce and
population) and the
exact alignment of
these points with the
center plaza or node
of commerce and
population .
The western
region of San
Cristobal Island
is definitely
suitable.
The eastern
region of San
Cristobal Island
is definitely
unsuitable.
The linear
function was
used. The
unsuitable
locations with a
minimum value
of 6 were
assigned a
membership
value of 0. Then
a membership
rating of 1 was
assigned to the
maximum value
of 5.
3e-Should be
developed near
or within the
existing urban
land use
ESA, Land
use land
cover layer
The layer is
comprised of land
use land cover area
measurements that
correspond to a
standard ESA
classification
system of
landscape types and
numerical codes.
The ESA's remotely
sensed data is
visualized at the 30
arc sec resolution.
This generalization
of the land use land
cover phenomena
across a 30 arc sec
cell area is a cause
of uncertainty.
The western
region of
Isabela,
northern region
of Santa Cruz
and portions of
northern San
Cristobal are
definitely
suitable.
The most
southern region
of Santa Cruz
and Isabela
Island are
unsuitable.
The linear
function was
used. The
unsuitable
locations with a
minimum value
of 10 were
assigned a
membership
value of 0. Then
a membership
rating of 1 was
assigned to the
maximum value
of 202.
8 9

3f-Should be
developed in
proximity to a
coastal port
Layer
digitized
from the
OSM
secondary
source
The layer is
comprised of
coastal route access
points in
association with
euclidean distance
measurements at 5
miles, 6 miles and
greater than 6
miles.
There is uncertainty
in the positional
accuracy of coastal
route access points
and the exact
alignment of these
points with the
primary roads.
The western
region of San
Cristóbal,
eastern region
of Isabela and
majority of
Santa Cruz
Island is
definitely
suitable.
The western
region of
Isabela, eastern
most region of
Santa Cruz and
eastern region
of San Cristobal
Island are
definitely
unsuitable.
The linear
function was
used. The
unsuitable
locations with a
minimum value
of 6 were
assigned a
membership
value of 0. Then
a membership
rating of 1 was
assigned to the
maximum value
of 5.
9 0

Appendix H. Analysis of the scenario two temperature
category of criteria.
Criteria Source Layer
Values in source
layer
Source of
Uncertainty
Suitable
Location(s)
Unsuitable
Location(s)
Membership
Function
4b-Avoid
regions that
contain
higher or
lower than
average
temperature
for the
majority, of
the year
WorldClim
Group,
Average
Annual
Temperature
layer
The layer is
comprised of
numerical
temperature
values in degrees
Celsius (values
converted to
Fahrenheit).
The raster layer is
visualized in 30 arc
sec resolution. This
representation of
average annual
temperature at the 30
arc sec scale is a
cause of uncertainty.
The eastern region
of Isabela,
southern region of
Santa Cruz and
western region of
San Cristobal
Island is definitely
suitable.
The western
region of
Isabela Island
is definitely
unsuitable.
The near
function was
used. A
membership
rating of 1 was
assigned to the
value of 73.4
degrees
Fahrenheit as the
midpoint and
membership
decreases on
either side of the
midpoint to an
assigned value of
0 for the value
62.6 and 95.0
degrees
Fahrenheit.
9 1

Abstract (if available)

Abstract Galapagos Island’s current agricultural system of monocropping, massive food imports, and a booming tourism sector has provided an increase in income for most galapaguenos that reside in the island but has been deemed unsustainable by the UNESCO World Heritage Organization. The tourism-driven urban development and monoculture system of food production have contributed to declines in water, wildlife habitat, soil quality, and an overall loss in biodiversity. This tourism sector growth along with a reduction in agroforestry production has reduced the income diversification potential for galapaguenos that reside in the islands and continues to threaten biodiversity. The most notable and critical of these global initiatives around biodiversity is goal seven of the global Millennium Development Goals (MDGs) of the United Nations (UN) which has since been translated into goal fifteen of the revised Sustainable Development Goals (SDGs). The goal seven of MDGs was targeted at ensuring environmental sustainability and parts of this goal were eventually folded into goal fifteen of the SDGs targeted at restoration and promotion of sustainable use of terrestrial ecosystems or life on land. These targets for goal fifteen have yet to be achieved. The scenario-based fuzzy modeling study was designed to support organizations focused on land use planning and management for agroforestry production and tourism development within the Galapagos Islands agricultural zone of San Cristobal, Santa Cruz, and Isabela utilizing a site suitability analysis framework. The framework was developed based on (1) contextual ecosystem requirements, (2) proximity to built environment infrastructure, and (3) availability of data. The framework implementation identified scenario 1: agroforestry production as being suitable across 20 percent of the study area or 12,386.40 acres and scenario 2: tourism development being 58 percent suitable within the study area or 35,920.56 acres.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Site suitability analysis for implementing tidal energy technology in southern California

PDF

Developing and implementing a GIS-based framework to identify optimal locations for clean water wells in sub-Saharan Africa

PDF

Identification and analysis of future land-use conflict in Mecklenburg County, North Carolina

PDF

Relocation bay: identifying a suitable site for the Tampa Bay Rays

PDF

Climate conservation priorities: using MCDA to identify future refuge for the Joshua Tree Forest in California

PDF

A critical assessment of the green sea turtle central west Pacific distinct population segment utilizing maxent modeling on nesting site locations

PDF

Community gardens for social capital: a site suitability analysis in Akron, Ohio

PDF

Geographic object based image analysis for utility scale photovoltaic site suitability studies

PDF

Onshore wind power systems (ONSWPS): a GIS-based tool for preliminary site-suitability analysis

PDF

Beyond visual line of sight commercial unmanned aircraft operations: site suitability for landing zone locations

PDF

Spatio-temporal analysis of Western snowy plover nesting at Vandenberg Air Force Base

PDF

Site location suitability analysis for a smart grid network

PDF

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

PDF

Analysis of future land use conflict with volcanic hazard zones: Mount Rainier, Washington

PDF

A site suitability analysis for an inland port to service the ports of Los Angeles and Long Beach

PDF

A cartographic exploration of census data on select housing challenges among California residents

PDF

Suitable California Opportunity Zone (COZ) locations for affordable housing development in the cities of Bakersfield, Los Angeles, and Palmdale

PDF

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

PDF

Wetland mapping and restoration decision making using remote sensing and spatial analysis: a case study at the Kawainui Marsh

PDF

Suitability analysis for wave energy farms off the coast of Southern California: an integrated site selection methodology