University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content ii

Spatio-Temporal Analysis of Western Snowy Plover Nesting
At Vandenberg Air Force Base

by

Regina Mary Butala

A Thesis Presented to the
Faculty of the USC Graduate School
University of Southern California
In Partial Fulfillment of the
Requirements for the Degree
Master of Science
(Geographic Information Science and Technology)

December 2019

iii

Copyright © 2019 by Regina Butala

iv

To Judith Ann Butala, my mother

v

Contents
List of Figures .............................................................................................................................. viii
List of Tables ................................................................................................................................. xi
Acknowledgements ....................................................................................................................... xii
List of Abbreviations ................................................................................................................... xiii
Abstract ........................................................................................................................................ xiv
Chapter 1 Introduction ................................................................................................................. 1
1.1 Western Snowy Plover Breeding Ecology ......................................................................2
1.2 Study Area .......................................................................................................................4
1.3 Current Monitoring and Management Efforts .................................................................8
1.3.1 Nesting and Distribution Monitoring .....................................................................8
1.3.2 Predator Management ..........................................................................................10
1.3.3 Habitat Restoration ...............................................................................................11
1.4 Research Objectives ......................................................................................................11
1.5 Document Outline .........................................................................................................12
Chapter 2 Related Research ....................................................................................................... 14
2.1 Nest Site Selection and Distribution .............................................................................14
2.2 Plover Habitat Restoration and Research ......................................................................16
2.3 Wrack Importance for Shorebirds .................................................................................19
Chapter 3 Data Collection and Management ............................................................................. 21
3.1 Field Data Collection Methods......................................................................................21
3.1.1 Data Sources .........................................................................................................21
3.1.2 Data Quality Assessment .....................................................................................25
3.2 Data Cleanup .................................................................................................................26
vi

3.2.1 Nest Data Processing ............................................................................................28
3.2.2 Wrack Data Processing ........................................................................................28
3.3 Conceptual Relational Database Design .......................................................................29
Chapter 4 Data Analysis............................................................................................................. 32
4.1 Plover Nesting Sites Analysis Methods ........................................................................32
4.1.1 Annual Hot Spot Analysis – 2D ...........................................................................33
4.1.2 Space-time Cube and Emerging Hot Spot Analysis – 3D ....................................35
4.1.3 Space-time Cube 3D Visualization ......................................................................40
4.2 Wrack and Nest Initiation Exploration ..........................................................................41
Chapter 5 Results ....................................................................................................................... 42
5.1 Annual Hot Spot Analysis .............................................................................................42
5.1.1 Initiated Hot Spot Distribution on WAL and SNO Restoration Areas ................43
5.1.2 Hatched Hot Spot Distribution on WAL and SNO Restoration Areas ................47
5.2 Emerging Hot Spot Analysis .........................................................................................47
5.2.1 North Beaches ......................................................................................................48
5.2.2 South Beaches ......................................................................................................51
5.3 Space-time Cube Hot Spot 3D Visualization ................................................................54
5.4 Wrack Abundance .........................................................................................................60
5.4.1 Initiated Nests Wrack Values ...............................................................................61
5.4.2 Hatched Nests Wrack Values ...............................................................................61
Chapter 6 Discussion and Conclusions ...................................................................................... 63
6.1 Result Overview ............................................................................................................64
6.1.1 Are there consistent temporal and spatial nesting hot spots in three categories:
nest initiation, successful clutch hatch and clutch failure due to predation? What
are the key areas for focusing future predator management and restoration work? ...64
vii

6.1.2 Which beaches of VAFB have hot spots of plover nesting and do these areas
correlate with recent habitat restoration activities? Have hot spot distributions
changed post restoration in these areas? .....................................................................65
6.1.3 Is there a correlation between high wrack abundance and nest initiation or
hatch? ..........................................................................................................................67
6.2 Management Recommendations and Future Work .......................................................68
6.3 Conclusions ...................................................................................................................69
References ..................................................................................................................................... 70
Appendix A. Extract of Plover Nesting Data Post Cleaning ........................................................ 74
Appendix B. Emerging Hot Spot Analysis Categories ................................................................. 75
Appendix C. WAL and SNO Restoration Areas Hatched HSA ................................................... 76
Appendix D. STC Hot Spot Visualization Hatched and Predated ................................................ 79

viii

List of Figures
Figure 1. Adult Western Snowy Plover .......................................................................................... 3
Figure 2. Study Area – VAFB Beaches .......................................................................................... 6
Figure 3. VAFB Beach Monitoring Segments and Transect Blocks .............................................. 7
Figure 4. Trends in annual number of nests initiated for North and South beach from
1994-2018 ........................................................................................................................... 9
Figure 5. North Beach Nest Distribution for MIN, SHN/SHS, and SAN. ................................... 10
Figure 6. Relative mean nest densities on Wall and Surf North in the contoured areas, the
adjacent beach immediately west of the contoured areas, and all South beach south
of Surf Open Area (control) from 2011-2018 ................................................................... 17
Figure 7. Data Cleansing Conceptual Model ................................................................................ 27
Figure 8. Nest Data and Wrack Data Cleansing and Integration Workflow ................................ 27
Figure 9. Conceptual Plover Relational Database ........................................................................ 31
Figure 10. Standard normal distribution of p-value and z-scores for 90%, 95%, and 99%
confidence levels ............................................................................................................... 33
Figure 11. Workflow of data preparation for the HSA. Annual data were separated into North
and South beaches, then into three categories: initiated, hatched and predated nests. ..... 34
Figure 12. Aggregation of data points into space-time bins ......................................................... 36
Figure 13. A space-time cube ....................................................................................................... 36
Figure 14. How Emerging Hot Spot Analysis Works. Shows a representation of an STC and
resulting EHSA ................................................................................................................. 39
Figure 15. Workflow Diagram of EHSA ...................................................................................... 40
Figure 16. Restoration Area Initiated Nest Points. ....................................................................... 43
Figure 17. Restoration Areas Initiated Nest Hot Spot Analysis 2004-2006 ................................. 44
Figure 18. Restoration Areas Initiated Nest Hot Spot Analysis 2007-2009. ................................ 45
Figure 19. Restoration Areas Initiated Nest Hot Spot Analysis 2010-2012. ................................ 45
Figure 20. Restoration Areas Initiated Nest Hot Spot Analysis 2013-2015. ................................ 46
ix

Figure 21. Restoration Areas Initiated Nest Hot Spot Analysis 2016-2018. ................................ 46
Figure 22. SHN – EHSA Results .................................................................................................. 50
Figure 23. SHS - EHSA Results ................................................................................................... 50
Figure 24. SAN - EHSA Results .................................................................................................. 51
Figure 25. WAL - EHSA Results ................................................................................................. 52
Figure 26. SNO – ESHA Results .................................................................................................. 53
Figure 27. SSO - ESHA Results ................................................................................................... 53
Figure 28. MIN - STC Initiated Nest Hot Spot Visualization. ..................................................... 55
Figure 29. SHN – STC Initiated Nest Hot Spot Visualization. .................................................... 55
Figure 30. SHS – STC Initiated Nest Hot Spot Visualization. ..................................................... 56
Figure 31. North SAN – STC Initiated Nest Hot Spot Visualization. .......................................... 56
Figure 32. South SAN – STC Initiated Nest Hot Spot Visualization. .......................................... 57
Figure 33. WAL – STC Initiated Nest Hot Spot Visualization. ................................................... 57
Figure 34. North SNO – STC Initiated Nest Hot Spot Visualization. .......................................... 58
Figure 35. South SNO – STC Initiated Nest Hot Spot Visualization. .......................................... 58
Figure 36. SSO – STC Initiated Nest Hot Spot Visualization. ..................................................... 59
Figure 37. North and South Beach wrack values indicating the percent of initiated and percent
hatched nests per wrack index value for 2012-2018. ........................................................ 61
Figure 38. WAL and SNO Restoration Areas Hatched HSA 2004-2006. .................................... 76
Figure 39. WAL and SNO Restoration Areas Hatched HSA 2007-2009. .................................... 76
Figure 40. WAL and SNO Restoration Areas Hatched HSA 2010-2012. .................................... 77
Figure 41. WAL and SNO Restoration Areas Hatched HSA 2013-2015. .................................... 77
Figure 42. WAL and SNO Restoration Areas Hatched HSA 2016-2018. .................................... 78
Figure 43. MIN – STC Hatched Nest Hot Spot Visualization. ..................................................... 79
Figure 44. MIN – STC Predated Nest Hot Spot Visualization. .................................................... 79
Figure 45. SHN – STC Hatched Nest Hot Spot Visualization. .................................................... 80
x

Figure 46. SHN – STC Predated Nest Hot Spot Visualization. .................................................... 80
Figure 47. SHS – STC Hatched Nest Hot Spot Visualization. ..................................................... 81
Figure 48. SHS – STC Predated Nest Hot Spot Visualization. .................................................... 81
Figure 49. North SAN – STC Hatched Nest Hot Spot Visualization. .......................................... 82
Figure 50. North SAN – STC Predated Nest Hot Spot Visualization. ......................................... 82
Figure 51. South SAN – STC Hatched Nest Hot Spot Visualization. .......................................... 83
Figure 52. South SAN – STC Predated Nest Hot Spot Visualization. ......................................... 83
Figure 53. WAL – STC Hatched Nest Hot Spot Visualization. ................................................... 84
Figure 54. WAL – STC Predated Nest Hot Spot Visualization. ................................................... 84
Figure 55. North SNO- STC Hatched Nest Hot Spot Visualization. ............................................ 85
Figure 56. North SNO – STC Predate Nest Hot Spot Visualization. ........................................... 85
Figure 57. South SNO – STC Hatched Nest Hot Spot Visualization. .......................................... 86
Figure 58. South SNO – STC Predated Nest Hot Spot Visualization. ......................................... 86
Figure 59. SSO – STC Hatched Nest Hot Spot Visualization. ..................................................... 87
Figure 60. SSO – STC Predated Nest Hot Spot Visualization. .................................................... 87

xi

List of Tables
Table 1. Summary of Data Received ............................................................................................ 22
Table 2. Extracts of Transect Survey Data Spreadsheet Predator and Wrack .............................. 23
Table 3. Extracts of Transect Survey Data Spreadsheet Avian Species ....................................... 23
Table 4. Extracts of Transect Survey Data Spreadsheet Plover Observations ............................. 24
Table 5. Nest Point Data Sample from 2002-2010 ....................................................................... 25
Table 6. Nest Point Data Sample from 2011-2018 ....................................................................... 25
Table 7. Extract of 2014 Wrack Data Week Formatting .............................................................. 29
Table 8. Hot Spot Analysis Tool Parameters ................................................................................ 35
Table 9. Space-time Cube Tool Parameter Settings for Analysis. ................................................ 37
Table 10. Emerging Hot Spot Analysis Parameters...................................................................... 40
Table 11. Seven hot spot categories detected ............................................................................... 48
Table 12. North Beach Hot Spot Categories ................................................................................. 49

xii

Acknowledgements
I am thankful for my family and my friend Lynne Hargett for their constant support through this
process. I would like to thank Point Blue Conservation Science and Vandenberg Air Force Base
for access to the 17-year dataset. Most of all, I am grateful for the decade I spent walking
hundreds of miles on the sand dunes to conserve and monitor the plovers.

xiii

List of Abbreviations
EHSA Emerging Hot Spot Analysis
GIS Geographic information system
GISci Geographic information science
HSA Hot Spot Analysis
MIN Minuteman Beach Segment
NetCDF Network Common Data Form
PCO Purisima Colony
PNO Purisima North Beach Segment
SAN San Antonio Monitoring Segment
SHN Shuman North Monitoring Segment
SHS Shuman South Monitoring Segment
SNO Surf North Monitoring Segment
SSO Surf South Monitoring Segment
SSI Spatial Sciences Institute
STC Space-time Cube
USC University of Southern California
USFWS United States Fish and Wildlife Service
VAFB Vandenberg Air Force Base
WAL Wall Monitoring Segment

xiv

Abstract
The population decline of Western snowy plover (Charadrius nivosus nivosus) and subsequent
listing as a threatened species by the U.S. Fish and Wildlife Service (USFWS) along the Pacific
Coast, is a result of poor reproductive success that is considered directly related to habitat loss
and nest predation. Habitat restoration and predator management are active key components to
the recovery plan of this species. Usually “good faith” restoration plans are carried out often
without the site-specific understanding of nest distribution and other factors that influence
nesting to focus these efforts. Bottom-up factors, such as food availability, may contribute to the
nest initiation by courting adult Western snowy plovers but these factors have not been directly
assessed during the breeding season. Habitat varies between breeding sites; therefore, it is
important to determine spatial patterns and regionally unique nest site selection for management
actions. The goal of this study is to better inform management regarding where future habitat
restoration or predator management activities need to be focused at Vandenberg Air Force Base.
This thesis looked at the spatial and temporal changes in Western snowy plover nesting
from 2002-2018 using Hot Spot Analysis to determine clustering of nest predation, initiation,
and success throughout the breeding site. Specific areas of habitat were identified as significant
hot spots in each nest category. These areas did not vary significantly year to year, however,
analyzing 17 years together summarized hot spot trends overall which pinpoint significant areas
where management actions should focus. Additionally, an exploratory analysis using habitat data
on wrack abundance was used to identify possible spatial correlation between this habitat
variable and nesting. The result of this analysis suggests no correlation between high wrack
abundance and nesting, rather it indicates that low wrack abundance was more prevalent than
high abundance during nest initiation.
1

Chapter 1 Introduction
Conservation management of threatened and endangered species populations under the
Endangered Species Act is constantly balancing the protection of wildlife with the allocation of
funding available that benefits the most species and habitats. Many resources are used to fund
and implement large management plans designed to aid in recovery of these species. Finding
“hot spots” or statistically significant clusters of biological data such as endangered species
locations is needed so that priority areas for management and restoration projects can be
distributed appropriately to support recovery goals (Trindade-Filho and Loyola 2011; Meyers et
al 1999). Therefore, research that examines nesting patterns, including their distribution and
related conditions in the environment that influence a species survival or reproductive success,
are crucial in allocating resources appropriately.
In the last several decades, habitat and predator management have been at the forefront of
the threatened Western snowy plover recovery plans in the coastal areas of California (USFWS
2007). Invasive plant species such as European beach grass (Ammophila spp.) have taken over
the coastal dunes throughout the plover’s breeding range altering habitat structure, reducing
available nesting areas, and impacting nest success (Muir and Colwell 2010; Zarnetske et al.
2010). Common predator populations such as coyote and raven have risen steeply in coastal
areas and have been directly linked to reduced Western snowy plover reproductive success (Page
et al. 1983; Neuman et al. 2004; Burrell and Colwell 2012).
Removing nest predator and invasive plant species (habitat restoration) are common
practices in most Western snowy plover management plans to increase reproductive success.
Many “good faith” restoration activities continue in breeding areas on the idea that if you restore
it, the species will use it (Ahlering and Faaborg 2006). However, some habitat remains
2

unoccupied by breeding plovers despite these measures (Robinette et al. 2017). Other factors
may be at work to influence where plovers place nests, including nesting densities or conspecific
attraction (Nelson 2007; Leja 2015; Patrick and Colwell 2017) and bottom-up environmental
changes such as algal wrack deposits that serve as food resources for nesting birds (Dugan 2003;
Lafferty 2013; Robinette et al. 2017).
This thesis analyzed and summarized the annual Western snowy plover nest initiation,
predation and hatching hot spots throughout Vandenberg Air Force Base breeding area. In
addition, the relationship between the spatial patterns of annual plover nesting to algal wrack
deposits or food resources was explored. The research goal was to better inform plover managers
of where and when to focus restoration and predator management efforts.
1.1 Western Snowy Plover Breeding Ecology
Western snowy plover (herein referred to as plover) is a threatened shorebird listed by the
USFWS under the Endangered Species Act on March of 1993. The population decline leading to
the plovers’ listing is a result of poor reproductive success due to habitat degradation, human
disturbance and predation of eggs and chicks (Page and Stenzel 1981; Page et al. 1991; USFWS
2007). The population’s range extends from southern Washington to Baja California, Mexico,
with the majority breeding from southern San Francisco Bay to southern Baja California (Page
and Stenzel 1981; Palacios et al.1994).
3

Figure 1. Adult Western Snowy Plover
The plover primarily nests in sparsely-vegetated dunes, sand spits, dune backed beaches,
river mouths, lagoons, and salt pans (Page and Stenzel 1981). Nesting begins in early March with
hatching continuing through mid-August. Pairs of birds often nest multiple times during the
breeding season with an ability to re-nest within two weeks after a failed nesting attempt
(Warriner et al. 1986; Page et al. 1991). After a failed nesting attempt, plovers may move
hundreds of kilometers to nest at another site or re-nest at the same location (Stenzel et al. 1994;
Powell et al. 1997). Clutches usually consist of three eggs and are incubated an average of 27
days by both sexes (Warriner et al. 1986). Once hatched, plover chicks leave the nest cup within
hours of hatching to look for food and are cared for by the male for an average of 28 days until
fledging. During this time, chicks crouch and hide from predators amongst the wrack and other
debris on the beach front between feeding on invertebrates and being brooded by the male.
Plovers rely on terrestrial invertebrates, foraging primarily in the intertidal zone, on the
wet sand, and within algal kelp deposits (Dugan et al. 2003; Schlacher et al. 2017). Plovers use
visual cues to capture prey on the surface and in the air by running and pecking at kelp flies
(Page et al. 1995).
4

1.2 Study Area
Vandenberg Air Force Base (VAFB), provides 13.8 miles of protected coastline for the
largest snowy plover breeding populations in California by closing beaches to the public during
nesting season from 1 March until 30 September and conducting extensive management.
However, limited recreational access is open throughout approximately 1.25 miles of breeding
habitat during the plover breeding season.
Management includes predator control, beach closures, focused research, invasive plant
removal, dune recontouring and nest monitoring. VAFB has conducted intensive monitoring of
the population since 2001 and more recently began an ongoing beach restoration project which
aims to provide additional suitable nesting habitat (Robinette et al. 2017). Due to the
management at VAFB, the relatively remote location of the breeding sites, and the fact that most
of the beaches are relatively undisturbed by human recreation activities; VAFB is an ideal site to
analyze nesting distribution, suitability, and factors impacting to nesting.
VAFB plover breeding occurs on three geographically isolated beach sectors referred to
as North, Purisima, and South beaches (Figure 2). These three sectors are further divided into
nine monitoring segments shown in Figure 3. North beaches stretch 6.2 miles from Minuteman
beach to Purisima Point and are characterized by wide sandy beach fronts with significant back
dune that extends in some areas one mile inland. This sector is further separated into four
monitoring segments: Minuteman (MIN), Shuman North (SHN), Shuman South (SHS), and San
Antonio (SAN). MIN is comprised of 1.1 miles of sandy beach with heavily vegetated dunes and
includes 0.25 mile “open area” which is open during breeding season to military personnel
recreation. SHN segment begins at Shuman Creek south 1.6 miles to No Name Creek. Habitat in
this segment is comprised of moderately vegetated back dune with extensive open sand sheets.
5

SHS extends 1.4 miles from No Name Creek south to San Antonio Creek. It is characterized by
narrow beach fronts with sand sheets surround by dense vegetation cover. SAN begins at San
Antonio Creek south 2.1 miles to north rocky Purisima Point. Habitat is comprised of wide sandy
beach fronts with extensive open sand sheets with sparse vegetation.
Purisima beaches incorporate bluff back pocket beaches and dune areas near Purisima
Point. Purisima beaches are comprised of two monitoring segments, Purisima North (PNO) and
Purisima Colony (PCO). PNO is characterized by sand sheets and sandy pocket beaches from the
south end of SAN 1.3 miles to Purisima Point. PCO includes a fenced California least tern South
beaches consist of five miles of continuous sandy coastline backed by dune habitat and steep
bluffs. South beaches are comprised of three monitoring segments, Wall (WAL), Surf North
(SNO), and Surf South (SSO). WAL section begins at the north end of Wall beach 1.3 miles
south to Santa Ynez River mouth. On the north end there is a 0.25 mile “open area” used for
military recreational access while the remaining portion is closed. SNO extends 1.8 miles south
from Santa Ynez River mouth and consists vegetated foredunes with narrow beach fronts. This
section contains 0.5 miles of “open area” used for public recreational access. SSO is a 1.9-mile
section of narrow sandy beach with vegetated and steep bluff backs. Figure 3 illustrates the
monitoring segments within the two beach sections along with the monitoring transect blocks
described in Chapter 3.

6

Figure 2. Study Area – VAFB Beaches

7

Figure 3. VAFB Beach Monitoring Segments and Transect Blocks
8

1.3 Current Monitoring and Management Efforts
Annual plover monitoring at VAFB began in 1993 to estimate the annual nesting
population and reproductive success. It was not until 2002 that a formalized protocol was
developed to track and monitor the breeding population. Point Blue Conservation Science is
currently working to document all nesting plovers on VAFB and determine annual trends in
population. Researchers are particularly focused on documenting how nesting plovers are
responding to restoration efforts as well as predator pressures.
1.3.1 Nesting and Distribution Monitoring
The VAFB plover population has been highly variable from year to year, with a mean
population size of 248 adults and mean nest number of 359 nests from 2000-2018 (Robinette et
al. 2018). USFWS measures reproductive success by the number of chicks fledged per male
plover. However, at VAFB it is not possible to track chicks to fledge due to inconsistent banding
efforts throughout the years. Therefore, managers at VAFB rely on tracking clutch hatch success
to understand the trends in reproductive success. Like plover population size variability, clutch
hatch success is also highly variable with a mean of 46% nests hatch per year. However, most
years are either extremely below or well above this average.
Plover nest distribution has varied over the years. The mean number of nests initiated on
North and South beaches is similar between 1994-2018 (Figure 4). Most recently, however, there
has been a significant increase in the number of nest initiation on South beaches since 2014. Nest
initiation increases often are due to years of high predations, such as the spike seen in 2004
where there was an increase in coyote nest takes and as a result an increase in re-nesting or triple
clutches.
9

Figure 4. Trends in annual number of nests initiated for North and South beach from 1994-2018.
Source: Robinette et al. 2018.

On the northern most section of North beaches (MIN-SHS), nest initiations have been
declining since 2006 despite of habitat availability (Figure 5). As a result, most of the nesting on
North beaches appears to be taking place in aggregated areas at SAN. One of the theories
presented by researchers is that predator pressures are pushing plovers to nest in denser clusters
and possibly regulating the size of the breeding population in that area. In 2011, a peregrine
falcon eyrie (nest site) was established after decades of the peregrines being extirpated from
VAFB (Robinette et al 2018). Peregrines are often seen hunting shorebirds (including plovers)
on North beaches and are potentially having an impact on nesting. Another theory is the result of
habitat loss at MIN, SHN, and SHS beaches.
10

Figure 5. North Beach Nest Distribution for MIN, SHN/SHS, and SAN. Source: Robinette et al.
2018.
1.3.2 Predator Management
Poor clutch hatch success is largely due to predation of nests by two main nest predators:
coyotes (Canis latrans) and common ravens (Corvus corvax). Common ravens have been a
common nest predator throughout the range of the plover. However, it was not until 2004 that
ravens were detected on VAFB and began taking plover nests on beach sections (Mantech 2009).
This is believed to be due the expansion of the raven population into central California coastal
territories (Boarman and Heinrich 1999) due to human activities which have provided food and
habitat for ravens (Camp and Knight 1993; Boarman et al. 2006; Kristan and Boarman 2007).
Ravens predated 10 percent of all initiated nests on VAFB in 2018, with the highest year on
record being 2017 where ravens took 25 percent of all nests (Robinette et al. 2018).
Coyotes are known to predate nests primarily on South beach sections with the highest
occurring in 2017 with 20% of all initiated nests taken by coyote (Robinette et al. 2017). Both
predators are sought out for lethal removal by predator management contractors on VAFB prior
to annual plover breeding season and continues until nesting is completed.
11

1.3.3 Habitat Restoration
VAFB initiated an extensive beach restoration management plan that aims in removing
invasive species European beach grass, ice plant, and Sydney golden wattle. These species have
a drastic impact on preferred nesting habitat of the plover by converting once open and dynamic
habitat into stabilized, densely vegetated monocultures (USFWS 2007). The goal of the VAFB
beach restoration plan is to transform these stabilized structures into natural plover habitat that
provides larger areas for nesting, chick rearing, and roosting (SRS Technologies 2005).
Prescribed burning followed by herbicide treatment has been the preferred treatment
strategy by restoration contractors. In 2007, restoration began on 800 acres of North beach at
SAN and PNO breeding sections. In 2009, 70 acres were restored north and south of the Santa
Ynez river mouth in the WAL and SNO breeding sections. Then from 2013 to 2015, this area
was also contoured using heavy equipment to breakdown unnatural dune structures formed by
the vegetation. In 2013, 180 acres were burned and treated in PCO section with annual follow-up
herbicide treatments to regrowth. In 2014, VAFB began yet another restoration project on North
beach in the sections of MIN. Thousands of pounds of chemical and millions of dollars have
been used over the course of these projects with so far very minor to no significant increase of
plover use for breeding (Robinette et al. 2018; Mantech 2018).
1.4 Research Objectives
Nesting occurs throughout the 10-mile coastline of VAFB with annual variation in
distribution which appears to be dependent on level of nest predation and habitat quality
(Robinette et al. 2017). Efforts to increase nesting success and population are focused on large
scale habitat restoration and predator management. However, there has been no work conducted
to determine where the most important areas or hot spots of plover nest initiation, clutch hatch
12

success, or nest predation occur to implement these activities. In addition, not much is known
about how other factors such as bottom-up influences of food availability within wrack deposits
may influence nest establishment. The goal of this study is to inform managers where important
nesting areas are and to provide additional information that can be used to redistribute
management actions (restoration, beach closures or predator removal) to be conducted in hot spot
areas. This will help in distributing funding wisely, refraining from “good faith” restoration in
lieu of intentional restoration.
The scope of this study was to provide answers to the following research questions using
17 years of plover nesting data at VAFB:
1. Are there consistent temporal and spatial nesting hot spots in three categories:
nest initiation, successful clutch hatch and clutch failure due to predation? What
areas are key spots for focusing future predator management and restoration
work?
2. Which beaches of VAFB have hot spots of plover nesting and do these areas
correlate with recent habitat restoration activities? Have hot spot distributions
changed post-restoration in these areas?
3. Is there a correlation between high wrack abundance and nest initiation or hatch?
I hypothesize that significant hot spots occur in central areas of the plover breeding areas
and they will be highly concentrated near river mouths. Further, I hypothesize that a larger
percent of nest initiations occurs during periods of high wrack abundance.
1.5 Document Outline
This study contains five additional chapters. Chapter Two begins with an overview of
related research about plover nest distribution, predator management and wrack importance, and
13

continues with an exploration of related literature regarding hots spot analysis and related field
data management. Chapter Three explores the field collection methodologies of the data used for
the analysis and how these methodologies can conflict with the analysts need. It demonstrates the
process of teasing the data into a useable form for this analysis and further demonstrates the need
for the use of a relational database. Chapter Four presents the methodology for data analysis.
Chapter Five presents the results and finally Chapter Six presents the implications of the results,
further recommendations, and future research suggestions.

14

Chapter 2 Related Research
Factors contributing to the nest site selection of the plover have been studied on several breeding
grounds within its range. None, however, have been conducted at VAFB. Studies that analyze
habitat mainly focus on microhabitat or specific substrate directly near the nest cup. Studies
regarding spatial distribution of nests have been conducted in many locations as it relates to
predators specifically, however no research was found analyzing the possible relationship
between wrack subsidies and nesting. Habitat restoration within breeding ranges is well
represented in the current literature. Impacts of restoration on plover nesting have been
investigated at a few sites. However, no previously published research was found on spatio-
temporal nest distribution using hot spot analysis. Below is a review of relevant studies related to
this thesis involving spatial distribution, nest site selection, a restoration review, and wrack
subsidies as it relates to shorebirds or plovers.
2.1 Nest Site Selection and Distribution
Nest aggregation and distribution of plover nests has been studied at the central and
northern most edge of its range. In Northern California, Patrick et al. (2017) found that
population density and nest aggregation had a strong correlation. At their site they found a large
portion of suitable habitat was left unoccupied while the population seemed to breed in
aggregated patterns and during years of higher populations plovers nested closer to one another.
This study highlighted the need to examine nesting patterns within the center of plover breeding
range where population size is larger to determine how the relationship between population size,
suitable habitat and the degree of nest aggregation. The main difference between the breeding
population used for this study and VAFB’s population, is that all their birds are color marked.
Meaning that they can distinguish each individual and as a result confirm all currently active
15

nests. Although the number of adult plovers can vary annually, the breeding population in
Northern California is relatively small (64) compared with VAFB population size (249) making
VAFB an ideal study site for nesting distribution.
Saalfeld et al. (2012) analyzed spatial distribution and nest site selection variables in the
inland population of plovers to determine nest characteristics involved in site selection to better
inform habitat management. This study used logistic regression and kernel density estimation to
identify habitat variables and to asses hot spots. They found that that over 57% of nests were
located within 100m of the nearest active nest site and nearly all nests were located adjacent to
an object such as a rock or plant. This suggests that nesting near an object may be chosen to
protect the eggs during extreme weather events. This conclusion is in alignment with the findings
of Page et al. (1985) who also analyzed nesting near objects and found a positive correlation
between success and object placement.
Fahy’s (2008) study of the breeding population at the Guadalupe Oil Fields just north of
VAFB analyzed both micro habitat variables and nest distribution over time. Fahy confirms the
above two studies’ findings about nest placement near objects but also looks at the spatial
distribution of nests. Analysis strategies focused on regression, generalized linear models and
nearest neighbor methodologies. Fahy found that nest distribution varies over years from random
to aggregated. The importance of Fahy’s research to this thesis is that the population studied was
just north of the VAFB population of snowy plovers and findings may be closely related to
regional habitat influences.
Eberhart-Phillips et al. (2016) analyzed the spatial distribution of the entire breeding
range of the snowy plover. They found that population growth in the southern regions were due
to predator management and that negative growth was linked to nest exclosures and harsh winter
16

conditions. This research also found that studies at the range-wide scale are misleading when
region-specific mechanisms such as varying climate and management practices are at work that
may impact the local breeding populations. They recommend that more studies be conducted on
regional scales rather than larger metapopulation analysis such as theirs to aid in conservation
and management of the species.
Another study was conducted at Mono Lake, a non-threatened inland population of the
plover which found that nesting density is directly related to predator pressures (Page et al.
1983). Although Page et al. (1983) point out that food distribution has an impact on nest
distribution, they do not directly look at the possible correlations between the two.
2.2 Plover Habitat Restoration and Research
Habitat restoration at VAFB is at the forefront of plover management. While no research
was conducted prior to carrying out the removal of invasive plants and dune contouring at
VAFB, follow-up monitoring has been ongoing since the restoration. Other breeding sites have
already documented response of the plover to restoration and most have found a favorable
response to restoration initially, however, if the areas are not consistently revisited for follow up
treatment and contouring, nesting begins to decline in those areas as native plants begin to
recolonize and dune systems are restored to natural slopes (Zarnetske et al. 2010). This has been
observed at VAFB as well by current biologists (Robinette et al. 2018). Average mean nest
density increased substantially following both contouring events (Figure 6) at WAL and SNO
and this was followed by a decrease in nest initiation beginning in 2016 when native vegetation
began taking over the recently barren ground created by the removing the dune structures
(Robinette et al. 2018). Powell and Collier (2000) recommend continued evaluation of the
restored nesting areas and cautions that it should not be based on the presence of nesting alone
17

but tracking of productivity over time. This conclusion is based on restoration of a nesting site in
San Diego county where factors contributing to increased use and reproductive success were
found to be related to habitat quality, but this led also to subsequent predation pressure in the
restored areas. Determining if there is any correlation in the distribution of initiated nest hot
spots pre- and post-restoration at VAFB is analyzed in Chapter 5.

Figure 6. Relative mean nest densities on Wall and Surf North in the contoured areas, the
adjacent beach immediately west of the contoured areas, and all South beach south of Surf Open
Area (control) from 2011-2018. The vertical line represents when dunes were contoured. Source:
Robinette et al. 2018
Zarnetske et al. (2010) analyzed plover response and non-target species response to
restoration. They found that plover nesting success following the reduction of European
beachgrass can occur initially but sustaining this increase may be possible only after repeated
treatments or contouring of dunes. Within the first year of restoration, there was increased
18

nesting, however in the following years, numbers of nests decreased unless the habitat was
actively managed yearly. It appears that plovers responded to the initial removal of vegetation
and the overall bare ground. Caution is given from this research that the single species approach
to restoration will increase target species population but may have long term potentially negative
impacts on other species and ecosystem functions.
Nelson (2007) found that breeding habitat may not be the limiting factor in plover
breeding population in Humboldt county. Conspecific attraction or social cues of plovers may be
more important to colonizing a breeding area rather than habitat restoration or removal of beach
grass alone. Therefore, the success of restoration that is focused on increasing plover
productivity and promoting plover recolonization is halted if the site itself is unoccupied by other
individuals. In addition, this study also found that plovers may choose a nest location not based
on habitat but based on the distance to active nests.
At the same study site, Leja (2015) looked at nest site selection and plover nest
establishment in areas where habitat restoration had occurred. This study found that in general,
plovers nested more frequently in restored habitats than unrestored habitats with 84 percent of
nesting occurring in these restored habitats. Restoration mimicking the natural dune processes
with yearly follow up were favored. Beach slope and width were the two physical habitat
characteristics that correlated with increase nesting at sites. Gentle slopes and wider beaches may
provide great visibility for the onset of predators, increasing nesting survival. The research
produced concrete management recommendations for future restoration such as no more than 4%
slope and beach width greater than 120 meters. In addition, the study found that the attraction to
nesting sites is heavily correlated with the presence conspecifics (other pairs of breeding
plovers). Social attraction may increase nest site selection and nesting densities in areas where
19

there are other plovers compared to physical habitat and restoration alone. Leja (2015) cautioned
managers that restoration alone may not attract plovers therefore careful consideration to
locations is vital for the success of any habitat restoration.
The overall take away from the literature regarding habitat restoration, is that human
induced restoration is successful in increasing plover nesting if it is continued over time and site
selection is based on the presences of neighboring conspecifics. Physical changes to habitat
alone, does not guarantee the use of the area by nesting birds.
2.3 Wrack Importance for Shorebirds
Sandy beaches around the globe rely on the input of macroalgae and seagrass
accumulations that support terrestrial macrofauna by increasing food availability and habitat
structure (Dugan et al. 2003; Ince et al. 2007; Schlacher et al. 2017). This ocean subsidy, or
wrack, supports terrestrial invertebrates which are often the center of the sandy beach food chain
(McLachlan and Brown 2006) by providing prey resources. Wrack has been directly correlated
with the distribution and abundance of shorebirds, including plovers, on sandy beaches
throughout the literature (Dugan et al. 2003; Brindock and Colwell 2011; Lafferty et al. 2013;
Dugan et al. 2003; Schlacher et al. 2017).
One of the most significant studies analyzing how macrofauna communities associated
with macrophyte wrack influence the abundance and spatial distribution of shorebirds is a local
Santa Barbara county study headed by Jenny Dugan et al. (2003). This study surveyed 15
beaches using transects to collect wrack samples and count shorebird species. They found that
increased wrack accumulations were positively correlated with macrofaunal abundance and the
number of plovers was found to have positive correlation to the amount of standing wrack.
20

There are few studies that analyze the correlation between wrack and plover abundance.
The studies that have been conducted occurred during the non-breeding season when plovers are
in wintering flocks. All show a positive correlation between wrack abundance and invertebrate
species abundance as a result a positive correlation between plover or shorebirds and wrack
abundance (Dugan et al. 2003; Lafferty et al. 2013). The assumption made in this thesis is that
large abundance of wrack provides and increased abundance of macro-invertebrates. In this
thesis wrack is explored as it relates to abundance during nest initiation.
21

Chapter 3 Data Collection and Management
The ability to share and conduct an analysis with data collected from researchers in the field
relies on the ability to “read” the data that are in a standardized format. Often data collection and
management are geared first to the field researchers needs and the results are not always user
friendly on the back end for future analysis. This chapter looks at the plover data sets acquired
for this study, the progression from data cleanup, data aggregating to processing, and to a final
dataset used for the analysis in this thesis. The last section discusses a conceptual model of a
relational database for all plover data for future archiving and analysis.
3.1 Field Data Collection Methods
Field collection of plover data has remained relatively consistent since 2002 when VAFB
began intensive monitoring of the breeding population. The same basic nest data have been
taken, however, over the years an increased amount of data have been collected in the field for
various research projects. Protocols for data collection in the field and the way they are archived
or the format they are entered, have been focused primarily from the field biologist perspective.
Real time data of nests status and estimations of hatch dates are calculated on the fly using the
data collected each day. This is done so that biologists can track nests throughout the incubation
period and birds throughout the breeding season. Therefore, formatting for data sharing is not
currently the priority in plover data management.
3.1.1 Data Sources
Spatial data, imagery, and spreadsheets of raw data were gathered for this thesis from
VAFB and Point Blue Conservation Science. A non-disclosure document was required to
22

procure the data from the Air Force and special permission to use data for this thesis was given
by Asset Management. Details of the data gathered are represented in Table 1.
Table 1. Summary of Data Received
Dataset File Type Data
Type
Details Source
2002-2010 Plover Nests Worksheet Point Nest location and
attribute data
VAFB
2011-2018 Plover Nests Worksheet Point
Plover Breeding Habitat Shapefile Polygon Boundary habitat
2011-2018 Wrack Data Worksheet Polygon Wrack index data Point
Blue Transect Block Shapefile Polygon Outlines transect blocks

3.1.1.1 Transect Block Surveys and Wrack Data
Transect block surveys began in 2012 by Point Blue Conservation Science and are
conducted each week during the breeding season. Each beach section described in Chapter One
is broken down further into transect blocks illustrated in Figure 3. Blocks are approximately 100-
300 meters in length along the coastal strand and extend the width of the beach from the current
high tide line during the survey day to the foredune. A transect in biological surveys, usually
refers to a linear segment where counting of plant and animal species are done on intervals along
a line in thin or thick bands (Montello and Sutton 2006). The transects in this study are arranged
as linear bands referred to as “blocks” and the areas within is where sampling occurs. The
transect block surveys are conducted on each beach section once per week throughout the plover
breeding season. Biologists walk down the beach counting the number plovers and other avian
species within each block, documenting the presence or evidence of terrestrial nest predators,
and recording the abundance of wrack present on the beach. Each transect block is given a single
value that estimates wrack abundance in the form of an index. This index ranks wrack presence
on a scale from zero to five: zero indicates there is no wrack present and five indicates heavy
deposits within the last high tide line.
23

The data files for these surveys contain several attributes: date, monitor, transect block
name, predator evidence, wrack index value, the number and type of plover or other avian
species present. Data are entered by field biologists in multiple tables which include all survey
years. Tables 2-4 include a sample of the transect data received for this project. Although data in
Tables 3 and 4 were not used for analysis in this thesis, they are shown here to demonstrate the
fragmentation of the data and need for long term storage without duplication in records. This is
discussed in Sections 3.1.2 Data Quality Assessment and 3.2 Conceptual Relational Database
Design.
Table 2. Extracts of Transect Survey Data Spreadsheet Predator and Wrack
Survey
Date
Monitor Transect
Block
Pig
Tracks
Coyote
Tracks
Raccoon
Tracks
Wrack
Index
8/6/2014 JKM Cobble No Yes No 1
8/6/2014 JKM Trex No Yes No 1
8/7/2014 LAH Border Trail No Yes No 0
8/7/2014 LAH Pebble No No No 2
8/7/2014 LAH GrandCyn No No No 2
3/10/2015 JKM SHSend Yes Yes No 1
3/10/2015 JKM Arrowhead Yes No Yes 0.5

Table 3. Extracts of Transect Survey Data Spreadsheet Avian Species
Survey
Date
Monitor Transect
Block
Avian
Species
Count Sex Age Behavior
8/6/2014 JKM Cobble WEGU 2 U U U
8/6/2014 JKM Trex CAGU 25 U U U
8/6/2014 JKM Trex WEGU 15 U U U
8/7/2014 LAH Pebble GBHE 1 U U tracks
8/7/2014 LAH GrandCyn GBHE 1 U U U
8/7/2014 LAH GrandCyn WEGU 1 U U U
3/10/2015 JKM SHSend MERL 1 U A Flushed plovers
3/10/2015 JKM Arrowhead RTHA 1 U A Flying
(U-unknown, A-Adult, WEGU-western gull, CAGU-California gull, GBHE-great blue
heron, MERL-merlin, RTHA-red-tailed hawk).

24

Table 4. Extracts of Transect Survey Data Spreadsheet Plover Observations
Survey
Date
Monitor Transect
Block
M F U Pair Chicks Juvenile
8/6/2014 JKM Cobble 1 2 0 1 0 0
8/6/2014 JKM Trex 4 3 1 2 0 1
8/7/2014 LAH Pebble 1 0 0 0 3 0
9/7/2014 LAH Pebble 2 1 1 1 0 0
(M/F- male or female plover, U- unknown sex, Pair-two plovers confirmed as breeding adults)
The main issue with the data in these tables is that many of the attributes repeat, such as
the survey data, monitor, and transect block. It is difficult to see from the different data files how
these data relate to each other. In addition, the wrack index data needed for the analysis in this
thesis are organized by survey date, not by transect block. Determining how to format this data to
make it useable was the first step in the process, followed by cleansing the data.
3.1.1.2 Nest Point Data
Each point represents a nest established in any given breeding season. Each beach section
is surveyed a minimum three days per week by plover biologists for the presence of active nest
sites and breeding birds. When a nest is located, the biologist records the GPS location, transect
block, and number of eggs. GPS accuracy ranges from 3-5 meters. Each nest is monitored until it
has hatched, predated or failed. An estimated initiation and fate date are added to the GPS
attribute file categorizing the nest as hatched, predated or failed. Failed nests are recorded when
a nest is washed out by tide, buried by sand due to wind, or nest abandonment. Nests are
recorded as predated when the nest fails prior to hatch date and there is evidence of predation at
the nest such as tracks or eggshell fragments.
Nest point data were acquired for monitoring years 2002 to 2018. Each nest was given a
number following the acronym for each beach section. The data were collected by two research
contractors who recorded and archived the data in different formats impacting data cohesiveness
for analysis. From 2002 to 2010 Mantech SRS collected data without the use of transect blocks
25

and stored by year in individual excel spreadsheets and received as annual shapefiles (Table 5).
Data spanning from 2011 to 2018 were collected by Point Blue Conservation Science and
received in one excel file (Table 6). It was during this period when transect blocks were created
and added to the attribute fields.
Table 5. Nest Point Data Sample from 2002-2010
Nest ID Latitude Longitude Initiation
Date
Eggs Hatch
Date
Fate Date Fate Cause
MM-01 34.8501970 120.6097910 4/5/2004 1 4/9/2004 Destroyed Tide
SA-104 34.7828950 120.6268440 7/4/2004 3 8/4/2004
SHN-02 34.8418870 120.6103910 4/5/2004 3 4/21/2004 Predated Raven
SHS-19 34.8213360 120.6149190 6/2/2004 1 6/4/2004 Predated Coyote
SN-024 34.6717140 120.6100130 4/7/2004 3 5/7/2004
W-02 34.6998020 120.6018700 3/29/2004 1 3/31/2004 Failed Unk
SS-083 34.6580670 120.6152170 6/2/2004 3 7/3/2004

Table 6. Nest Point Data Sample from 2011-2018
Nest ID N_UTM84 S_UTM84 Transect
Block
Eggs Initiation
Date
Fate
Date
Fate
11MIN001 3859182 718654 Alligator 1 5/10/2011 5/10/2011 Hatched
11SHN006 3857952 718628 Scaffolding 2 4/25/2011 5/5/2011 Abandoned
11SHS016 3855970 718149 SHSstart 3 5/6/2011 5/20/2011 Tide
11SAN021 3852201 717424 Stix 3 4/10/2011 4/28/2011 Raven
11SNO047 3840279 719242 NSurfopen 2 5/25/2011 6/21/2011 Hatched
11SSO027 3838207 718683 Squid 1 5/01/2011 5/24/2011 Coyote
11WAL042 3841770 719666 Cigar 3 5/25/2011 6/26/2011 Hatched
3.1.2 Data Quality Assessment
Most of the data received for this thesis required extensive processing prior to analysis
due to poor data quality and inadequate formats. The 2002-2018 nest data was in two formats
due to two organizations collecting and archiving data differently. These two datasets,
represented in Table 5 and 6, contained redundant data in different representations and needed to
be standardized and consolidated for integration into one file. Attribute headings in these tables
are inconsistent, nest identification numbers differ, nest coordinates are in different projections
and there is no transect block information in 2002-2010 dataset (Table 5). Prior to integrating
26

these files, all data needed to be “cleaned” which is a process using strategies to check values, fix
data anomalies, remove duplicates and validate data.
3.2 Data Cleanup
The process of data cleansing targets errors in the data or anomalies in the data that are
determined by the specific application of use or analysis and requires an expert in the field the
data is taken (Muller and Freytag 2003). It is especially needed when there are two data sources
that require integration (Rahm and Do 2000; Kandel et al. 2011). In this case, it would need to be
a biologist involved with plover conservation that understands the vernacular within plover
ecology thus the data cleansing was conducted by myself. While many large datasets use
programming, frameworks designed to clean data (Muller and Freytag 2003; Lee et al 2000),
plover data cleaning strategies were conducted for this thesis using excel filters, pivot tables, and
sorting functions. The outcome was an integrated plover nest data source applicable for use in
this thesis and capable of being uploaded to ArcGIS for analysis.
Figure 7 represents the conceptual model of data cleansing used in this thesis provided by
Lee et al. (2000). The first step was to sort through both datasets and clear up any duplicate
records, differences in spelling, categorical data values and attribute headers in all datasets.
Following this data cleansing model by Lee et al. 2003, both nest datasets and the transect data
underwent an enormous restructuring and manipulation to integrate all into one useable file for
analysis. Figure 8 depicts the model for data cleansing for this thesis and specific steps are
described in Sections 3.2.1 and 3.2.2.

27

Figure 7. Data Cleansing Conceptual Model. Source: Lee et al. 2000.

Figure 8. Nest Data and Wrack Data Cleansing and Integration Workflow
28

3.2.1 Nest Data Processing
The data formatting used here followed the basic structure shown in Table 6; therefore
the 2002-2010 data underwent the most manipulation to fit into the basic format of the 2011-
2018 dataset. First the projection of the 2002-2010 data set was changed in ArcPro and exported
into another excel file for further processing. Each of the nest names was converted into the
“Year – Beach Section – Nest number (18SAN052)”. Categorical values for nest fate were
examined and put into two fields “Fate” and “Fate Date”. These values were sorted to remove
spelling errors and inconsistent nomenclature. Data attributes were added to the 2002-2010 data
to match the 2011-2018 data. These attributes included: year, beach segment and transect blocks.
Finally, the nests were given a week number that refers to the calendar week of the specific year
in which it was initiated. The week number serves as a key to be used to append the wrack data
file.
3.2.2 Wrack Data Processing
Prior to using the transect data files (see Tables 2-4) for analysis, three steps of data
cleaning and integration were required. First, since this dataset included information on
shorebirds and predator observations, in addition to wrack data, for each transect, these fields
were removed to create a file that contained only wrack data. Then new data variables were
added: week number, year, beach section, and monitoring segment. By adding week numbers
and removing the individual survey dates, this data can “speak” to the nest data described above
by linking it with week numbers. In addition, repeated transect blocks were eliminated so that
each year had only one row for each transect block. Week numbers became columns and wrack
values were added to the columns along each transect block row (Table 7). Next this information
was used to assign a wrack value to each nest initiated in corresponding transect blocks. This
29

value was appended to the nest point data file. Appendix A contains an extract of the new
integrated datafile which includes both nest information and wrack value assignment per nest.
Table 7. Extract of 2014 Wrack Data Week Formatting
Year Beach
Section
Monitoring
Segment
Transect
Block
Week
10
Week
11
Week
12
Week
13
Week
14
Week
15
2014 North SHS OJ 4.0 2.0 1.0 1.0 0.5 1.0
2014 North SHS Bottlebuoy 3.5 2.0 1.0 1.0 1.0 0.5
2014 North SHS Mole 3.0 1.0 0.5 1.0 0.5 0.5
2014 North SHS Bottlelog 4.0 1.0 0.5 0.5 0.5 0.5
2014 North SHS Trilog 3.0 1.0 1.0 1.0 1.0 0.5
2014 North SHS SHSend 2.0 2.0 1.0 1.0 1.0 1.0
2014 North SAN Arrowhead 1.5 1.0 1.0 1.0 1.5 1.0
2014 North SAN L13buoy 1.0 1.0 1.0 1.5 2.5 1.0
2014 North SAN Niceperch 1.0 1.0 1.0 1.5 2.5 1.0
3.3 Conceptual Relational Database Design
A relational database is a set of tables containing data that are related to each other. Each
row in a table is labeled with a unique id that is called a “key” that links each table to one
another and determines their relationship to each other. Data consistency is maintained in the
relational database structure and allows for multiple users to access the same data in different
formats or queries. The benefit of this type of data structure is that less time is needed for data
management and data can be used for multiple analyses.
The complete plover nest data assembled through data cleaning and integration required
an extensive amount of time to create. If at the onset, these data had been stored in a relational
database, simple queries could have been performed to end up with the complete plover nest
dataset shown in Appendix A. Although current plover biologists use a Microsoft Access
database, the design of the database was not available for this project and not appropriate or set
up for the type of queries needed to aggregate data for this thesis. The structure of the Microsoft
Access database was designed for the field biologists to enter data using forms and to develop
30

queries for daily monitoring such as lists of nests that are due to hatch for that survey day. This
“on the fly” data query capability is ideal for the field biologist but is not ideal for the data
analyst. This Access database is also not complete, as it only includes data from 2011 to 2018
(the years when Point Blue had the monitoring contract). As a result, data were exported from
this database and provided by Point Blue Conservation in Excel spreadsheets. Ideally, a
relational database containing all years of nest data, from 2002 to 2018, would be set up so that
information would not be duplicated, and a vast array of queries could be run to allow for various
ways of aggregating the data for different analyses.
While processing data for use in this thesis, it became apparent that the creation of a
relational database that could be used to extract the data tables needed for analysis is greatly
needed. The effort used to deconstruct, clean, and integrate all the data used in this analysis
would have been significantly reduced if there had been one location where these data were
archived. It is now possible to create a plover relational database for future use after having
processed all the historic and current data from both VAFB, ManTech, and Point Blue
Conservation Science. Although this thesis did not create or populate a relational database,
Figure 8 presents the proposed conceptual database design diagram of a relational database that
would suit this type of data and further analysis. There is no duplicate data within this database
and each category is kept separate in order to relate or query information together.

31

Figure 9. Conceptual Plover Relational Database
A large part of the time spent on this thesis work was manipulation and cleansing of
existing data from multiple sources. Once the data were in an appropriate format, analysis could
begin using the complete and integrated 17-year nest data including wrack information. The
following chapters describe the analysis and results.

32

Chapter 4 Data Analysis
The purpose of this study is to assess the beaches of VAFB for plover nesting hot spots and make
preliminary assessments of how wrack impacts nesting sites. The 17-year dataset of nest
locations on VAFB provide an opportunity to analyze both spatial and temporal patterns of areas
that are consistently used for nesting, to identify where successful nesting occurs and to
determine locations with hot spots of nest predation. First, this chapter describes the different
variables used in the spatio-temporal analysis using hot spot analysis, space-time tools and
emerging hot spot analysis in ArcGIS Pro. The second part of the chapter describes the wrack
data exploration to determine any correlation between wrack and nest initiation.
4.1 Plover Nesting Sites Analysis Methods
To analyze VAFB nesting, data from North and South beach were analyzed separately
due to the geographic isolation of each segment and habitat differences. Each year of nest points
was separated into three categories for both North and South beach: initiated, hatched, and
predated nests. Two methods were used to identify spatio-temporal hotspots. The first analyzed
each plover nesting year separately in two-dimensional analysis. The second used a three-
dimensional data structure, a space-time cube, to organize all 17 years of plover nest data,
followed by an emerging hot spot analysis using the input cube. The parameters chosen for these
tools are crucial to the way these data were analyzed and dependent on the question being asked.
In this analysis, parameters were used based on the plover breeding ecology where possible and
they are described in each section below.
33

4.1.1 Annual Hot Spot Analysis – 2D
Hot spot analysis (HSA) is a test for randomness in data and identifies locations of
statistically significant hot or cold spots by, first, aggregating point locations into polygons or
grids, called features, generally weighted simply by the number of points in each area. It uses the
Getis-Ord Gi* statistic to assess the weight of each feature within the context of the weights of
features in its neighborhood (Getis and Ord 1992) and against the average weights of the study
area. If the neighborhood has a weighted value that is significantly higher than that of the study
area, then the feature is a hot spot. If the neighborhood value is significantly lower than that of
the study area, then it is a cold spot.
The Gi* statistic in ArcGIS Pro, returns a number for each feature in the dataset as a z-
score and p-value. Positive z-scores mean that it is an area of intense clustering or a hot spot.
Negative z-score values indicate low clustering or cold spots. Lower p-values indicate high
confidence that the pattern is different from random (Figure 10).

Figure 10. Standard normal distribution of p-value and z-scores for 90%, 95%, and 99%
confidence levels. Source: Esri 2018a.

34

Figure 11 shows the workflow of data preparation for the annual HSA. Each year of the
full 17-year plover nest data was extracted and then separated into the two beaches and those
were again separated into the three nest categories. The result is 102 separate shapefiles
disaggregated by year, beach and category. Each of these shapefiles was subjected to an HSA,
resulting in 102 hot spot shapefiles.

Figure 11. Workflow of data preparation for the HSA. Annual data were separated into North
and South beaches, then into three categories: initiated, hatched and predated nests.

Two distance parameters are required when running the hot spot analysis tool: distance
band and the size of the aggregation grid. Parameters were set using plover behavior and nesting
preferences cited in current literature (Table 8). The fixed distance band or neighborhood is the
distance the tool will use to determine which features are neighbors of each feature of interest.
This was set at 100m based on published literature that shows nearest nest neighbors are
generally located within 75-100m of the nearest active nest site (Saafeld 2012; Patrick et al.
2018). In addition, this is the distance of the average transect block and the average distance a
plover would be disturbed off a nest when a biologist was sighted (Butala unpublished data).
This distance (75-100m) does not represent territory size; territory size is the distance
surrounding an active nest that a plover will defend as described below.
35

Table 8. Hot Spot Analysis Tool Parameters
HSA Paremeter Description Setting Reason
Fixed Distance Band Neighborhood 100m
Plover nest average nearest
neighbor (Patrick et al. 2018)
Grid Shape Shape Hexagon
Less distortion, suited for
neighborhood analyes
Aggregating Grid Size
Size of the
Hexagon
20m
Plover average nesting
territory (Fahy 2008)

The parameter for the shape of the aggregation grid used was hexagons. There were two
choices for grid shape: fishnet (square) or hexagon. The hexagon grid was chosen rather than a
traditional square grid because it reduces edge effects, there is less distortion, and conducting
neighbor analysis between hexagons is more straightforward as the centroid of each neighbor is
equidistant from other hexagon neighbors (Birch et al. 2007). In addition, the study site has
irregular shapes along the coast and the hexagon grid fit the irregular pattern better, reducing
edge effects. The dimension used was 20m to reflect the plover nest territory size (Fahy 2008).
These parameters remained consistent when running all 102 models. All possible nesting area
was included which extends to all sandy beaches of North and South beach.
While it did not form the basis of the final analysis in this study, the results of the annual
HSA were useful in visualizing pre- and post-restoration nesting hot spots at WAL and SNO.
Because restoration in these areas was based on increasing nesting and success, only initiations
and hatch categories were analyzed. Maps of the two restoration sites were created from 2003-
2018 initiated and hatched nesting hot spots.
4.1.2 Space-time Cube and Emerging Hot Spot Analysis – 3D
A space-time cube (STC) is created using an input data layer of time-stamped points that
is restructured into a Network Common Data Form (NetCDF) data cube file using “create space-
time cube tool” in the Space Time Pattern Mining toolbox in ArcGIS. NetCDF is a data format
36

for multidimensional scientific data that is structured so that it is easy to access and display
selected attributes through a dimension such as time. An STC is made from NetCDF data for a
specific attribute dimension such that time stamped features are aggregated into space-time bins
(Esri 2019) where the x and y (horizontal) dimensions represent space and the t (vertical)
dimension represents time (Figure 12). Each bin has a fixed position in space (x,y) and in time
(t). The temporal data are set up with time-steps while the spatial extent of the cube is
represented by rows and columns (Figure 13) in a gridded pattern. In this analysis, the time-
stamped points used as input are nest points from 2002-2018 shapefile.

Figure 12. Aggregation of data points into space-time bins. Source: Esri 2018a.

Figure 13. A space-time cube. Source: Esri 2018a.

37

Three separate STC’s were created from the VAFB 2002-2018 plover nest data: all nests,
successful clutch hatch and predated nests. There are several parameters that are optional for this
tool, not all were used for this analysis. Parameters specific for this analysis are shown in
Table 9; these include time-step interval and distance interval (size of bins). All other parameters
were left at default or not used when tool was run. The time-step interval is used to determine
how to partition aggregated points across time (years, months, weeks, days or hours). This was
set at one year while the distance interval or size of the hexagon bin was set at 50m. Initially this
distance was set at 20m to mimic the setting for annual HSA which represents a plover nest
territory size (Fahy 2008). However, this proved to be too small and created a high frequency of
zero counts in bins which made it difficult to detect trends (Esri 2018a). On the other hand, if it
is set too high, the underlying pattern will be lost. After experimenting with this setting, 50m was
determined to fit the best with the data. Nest points that share the same space and time-step
interval were aggregated by the tool into hexagon bins with a one-year time step creating 17
time-steps in the plover cube. Although the nest data is limited to the breeding season only from
March-September annually, the tool recognizes the other months as null values. The three space-
time cubes generated were then used in EHSA to create hot spot maps and STC hot spot
visualizations to analyze the distribution of hot spots annually.
Table 9. Space-time Cube Tool Parameter Settings for Analysis.
STC Parameter Setting Description Reasoning
Time-step Interval 1 year Aggregates points across time Study is analyzing yearly plover
nesting
Size of Grid 50m How large the space-time
bins
Closest to 20m territory without
losing data
Bin Shape Hexagon Shape of Bin Less distortion, better when using
neighborhood analyses

38

The emerging hot spot analysis (EHSA) tool is an offshoot of hot spot analysis that uses
an STC to calculate the Getis-Ord Gi* statistic to determine if each bin is a statistically
significant hot or cold spot. From this evaluation, each bin is then given a z-score and a p-value.
Hot or cold spots are a result of larger positive or negative z-scores. Lower p-values indicate a
higher confidence that the pattern is not resulting from random chance.
Once hot spots are identified with the Getis-Ord Gi*, the Mann-Kendall statistic
evaluates hot and cold spots through time within each stack of bins in the space-time cube (Mann
1945, Kendall and Gibbons 1990). The bin value in a stack is compared to the bin values at
temporally adjacent positions. A +1, 0 or -1 value is assigned to each time period/bin. For
example, if the first bin is smaller than the second it gives a +1 score. If the first bin value is
larger than the second, it gives a score of -1. If there is no difference a value of zero is given. The
results of each pair of time periods are summed and statistical significance is determined by
evaluating this sum of the stack compared to the expected sum of zero.
As illustrated in Figure 14, combining the hotspot and trend z-scores and p-values for
each stack of bins, the EHSA tool classifies each location as one of 17 categories (Esri 2018b).
Since the full set of 17 categories is not relevant in this analysis, the description of each category
is provided in Appendix B for reference. The description of the seven categories relevant to this
study as provided in the next chapter.
39

Figure 14. How Emerging Hot Spot Analysis Works. Shows a representation of an STC and
resulting EHSA. Red represents hot spot categories while blue represents cold spot categories.
No color indicates that there is no pattern detected. Source: Esri 2018b.

The specific parameters used for this EHSA analysis were neighborhood distance,
neighborhood timestep, conceptualization of spatial relationships and polygon mask or area of
analysis (Table 10). The neighborhood distance, the distance used to determine neighboring bins,
used in this analysis was 75m which is based on the plover nesting nearest neighbor in the
literature (Saafield 2006; Patrick et al. 2018). The polygon mask that was used was the shapefile
of all possible plover breeding areas on VAFB. The neighborhood timestep was set at 1 and the
spatial relationship was set at fixed distance. The time-step interval determines which features
are analyzed together to assess space-time clustering. The conceptualization of spatial
relationships parameter determines how spatial relationships among bins are defined. In fixed
distance, each bin is analyzed within the context of the neighbor bins, those outside the
neighborhood distance have no influence on the target bin’s value those inside the distance exert
influence on the target bin.
40

Table 10. Emerging Hot Spot Analysis Parameters
EHSA Parameter Description Setting Reasoning
Neighborhood Distance The extent of the analysis
neighborhood.
75m Plover nest average nearest
neighbor 75-100m (Patrick
et al. 2018)
Conceptualization of
Spatial Relationships
Defines spatial
relationships among bins.
Those outside
neighborhood distance
receive weight of zero.
Fixed Distance Keeps the analysis within
the neighborhood distance
setting. Those outside are
not included.
Polygon Mask Defines the analysis study
area.
Breeding
Beach Polygon
This is the study area
Grid Shape Determines shape of the
aggregation bins.
Hexagon Less distortion, better when
using neighborhood analyses

Figure 15 shows the workflow from three STC and the analysis extent set in ArcGIS Pro
for North and South beaches. Six shapefiles were created then displayed on maps broken into
beach and beach segments for analysis, resulting six maps.

Figure 15. Workflow Diagram of EHSA
4.1.3 Space-time Cube 3D Visualization
After the EHSA, is run the results of the hot spot analysis are stored in the NetCDF cube
and can be visualized in 3D using the Visualize Space Time Cube tool. There are several options
41

for choosing a display theme; for this analysis, displaying the hot and cold spot results was used.
This visualization helps to understand the structure of the STC and allows exploration of the
results of the EHSA. This analysis used the 3D visualization to analyze different locations and
years to understand locations that did not indicate hot spots using the ESHA and the ability to
look at yearly trends. Thus, it is possible to explore each year in each location by moving
through the bin stacks. Maps show a general view but mainly this tool supports managers to
explore the data and investigate areas, such as sand sheets, where hot spots appear after years of
no hot spots. Several maps were created to present the 3D stacks in each beach section; however,
this tool is more useful as an interactive exploratory tool. Therefore, analysis occurred
interactively in ArcGIS Pro and results are discussed and illustrated in Chapter 5.
4.2 Wrack and Nest Initiation Exploration
An initial exploration on wrack transect data was conducted to gain insight on how wrack
abundance may impact nesting and help guide future work beyond this study. This initial
analysis was designed to determine if more nest initiations or successful clutch hatches occur
during periods of high wrack indices (abundance). The final 2002-2018 plover nest shapefile
(excerpt in Appendix A) was used in Excel to conduct a simple pivot table sort. Nests were
separated into two categories, all nests initiated and clutch hatch success, then sorted to
determine the number of nests initiated and hatched during each wrack value (0-5). The results
provide a quick look at which wrack category is more prevalent for 2012-2018 plover nest
initiation and successful hatches.

42

Chapter 5 Results
The results from the HSA, the EHSA and the STC hot spot visualization provide useful
information regarding the nesting trends of the plover and where key nesting hot spots are
located at VAFB. First, the results of the annual HSA are presented. Then the results of the
EHSA are reviewed followed by the STC hot spot visualization in 3D. The final section presents
the results from the nesting wrack analysis.
5.1 Annual Hot Spot Analysis
While comparing the annual HSA results, it was difficult to distinguish variations
between years given there were 102 maps to compare. It became clear that visualization of the
STC hot spots through the EHSA allowed for easier temporal analysis of the data. However, the
annual HSA is useful for looking at specific years in specific locations. In this section, the WAL
and SNO restoration areas are highlighted for evidence of changes in nest initiation and hatched
hot spots from 2003-2018, pre- and post-restoration. Restoration began in the winter of 2009
with completion in 2014/2015. Aerial imagery shown in all figures is from 2018 (post-
restoration), therefore visual representation of actual topography of beach is incorrect when
looking at past years data projections.
Figure 16 shows the raw nest point distribution pre- and post-restoration. Prior to
restoration (2002-2008), nest distribution was exclusively found on the coastal area in front of
the foredune, while post restoration (2009-2018), nest distribution spread to the east behind the
foredunes.

43

Figure 16. Restoration Area Initiated Nest Points. Pre-restoration (left), 2002-2008, nest points
initiated showing nest distribution primarily on the coast and along the river mouth in WAL and
SNO. Post-restoration (right), 2009-2018, nest points initiated showing nest distribution along
the coast and further to the eastern boundary.
5.1.1 Initiated Hot Spot Distribution on WAL and SNO Restoration Areas
Figures 17 through 21 show the initiated nest hot spot distribution for WAL and SNO
from 2002-2018. Initiated hot spot distribution on WAL remained consistent in most years pre-
and post-restoration on the southern edge of the restoration area and sporadic on the beach front.
The southern area is subjected to yearly influence of the Santa Ynez river mouth that may
influence and cause fluctuation in habitat quality from year to year. This hot spot did not change
significantly over the years after restoration, however, a new hot spot appeared at the north end
after restoration efforts (Figure 21).
44

SNO restoration did not show any hots spots until post-restoration years, 2014, 2016, and
2017 (Figure 20 and 21). In 2004, a large cold spot was detected along the eastern edge of both
restoration areas (Figure 17). Incidentally, this is where the thickest area of invasive vegetation
(beachgrass and golden wattle) was present before its complete removal in 2009. No other cold
spots were detected in all years.

Figure 17. Restoration Areas Initiated Nest Hot Spot Analysis 2004-2006. This figure shows
three years of nest initiation hot spots in the WAL and SNO restoration areas.
45

Figure 18. Restoration Areas Initiated Nest Hot Spot Analysis 2007-2009.
.
Figure 19. Restoration Areas Initiated Nest Hot Spot Analysis 2010-2012.
46

Figure 20. Restoration Areas Initiated Nest Hot Spot Analysis 2013-2015.

Figure 21. Restoration Areas Initiated Nest Hot Spot Analysis 2016-2018.
47

5.1.2 Hatched Hot Spot Distribution on WAL and SNO Restoration Areas
Hatched hot spot distribution followed a similar pattern as the initiation hot spots. At
SNO for all years prior to restoration, there were no hot spots detected except for northern and
coastal areas. The northern part of the restoration area, like WAL southern area, is impacted by
the Santa Ynez river with year to year habitat quality fluctuation likely impacting nesting. In
2014, the largest hot spot occurred in the center of the restoration area. Followed by 2016-2017
hot spots in that same area. Figures for hatched HSA for WAL and SNO restoration areas are in
Appendix C.
5.2 Emerging Hot Spot Analysis
Space-time emerging hot spot analysis was run on three categories of plover nest data: all
initiated nests, successful clutch hatches, and predated nests. Only statistically significant hot
spot and non-significant trends were found on the beach segments; no statistically significant
cold spots were detected. Hot spot patterns were more prevalent on the southern region of North
beach and on the northern region of South beach, both nearby river systems. Across all three
categories, the analysis identified only seven hot spot pattern categories: no pattern detected,
historical, new, consecutive, intensifying, persistent, and sporadic hot spots. Definitions of these
seven hot spot patterns are shown in Table 11 (See Appendix B for all hot spot category
definitions).
48

Table 11. Seven hot spot categories detected
Pattern Name Definition
No Pattern Detected There were not hot or cold spot patterns found in area
New Hot Spot A location that has never been a hot spot before but is a hot spot for
the final time step
Consecutive Hot Spot A location with an uninterrupted run of hot spot bins in the final time-
step intervals. Less than 90 percent of all bins are hot spots and the
location has never been a hot spot prior to the final run.
Intensifying Hot Spot A location that has been a hot spot for 90 percent of the time-step
intervals including the final interval. The intensity of clustering of
high counts in each time step is increasing overall and the increase is
statistically significant.
Persistent Hot Spot A location that has been a hot spot for 90 percent of the time-step
intervals with no apparent trend indicating an increase or decrease in
the intensity of clustering over time.
Sporadic Hot Spot Areas that are hots spots at one time-step interval then not a hot spot.
Less than 90 percent of the time-step intervals have been hot spots
and never have they been cold spots.
Historical Hot Spot At least ninety percent of the time-step intervals have been hot spots,
but the most recent time period is not hot.
5.2.1 North Beaches
Most of the North beach area did not show statistically significant trends. The ESHA
detected seven hot spot pattern categories, there were no statistically significant cold spots found
on any of the beach segments. Most hot spot categories occurred on SAN segment around San
Antonio river mouth. There were hot spots detected in each nest category in all beach segments
except for MIN (Table 12). There is no figure representing MIN beach segment due to no results
found.
49

Table 12. North Beach Hot Spot Categories
New Historical Consecutive Intensifying Persistent Sporadic
Initiated
SAN 7 - 17 5 25 117
SHN - - - - - 5
SHS 2 - 1 1 - 47
Hatched
SAN 5 1 - - 4 72
SHN - - - - - 1
SHS - - - - - 17
Predated
SAN 7 - 10 - - 93
SHN - - - - - 8
SHS 1 - 2 - - 21

The spatial distribution of North beach hot spots is shown in Figure 22 through 24 for
SHN, SHS, and SAN. SHN had the least number of hot spots detected, with only a few sporadic
hot spots in all three nesting categories which occur on the southern section. SHS resulted in a
new hot spot for both nest initiations and predations in the mid to southern section. Sporadic hot
spots occur on the extreme south end of SHS for both hatched and predated nests, where in the
initiated nest results, they occur throughout. There was one consecutive hot spot in the same
transect block area as the new hot spots in the initiated nest category at SHS. SAN segment
EHSA resulted in more hot spots than SHN or SHS, as shown in Figure 24. Consecutive hot
spots occur on SAN throughout the segment with one detected on the far east of the southern
sand sheet. Most of the significant hot spots occur north and south of the San Antonio river
mouth and sandspit.

50

Figure 22. SHN – EHSA Results

Figure 23. SHS - EHSA Results
51

Figure 24. SAN - EHSA Results
5.2.2 South Beaches
The EHSA detected three hot spot pattern categories on South beach; there were no
statistically significant cold spots (Table 8). The spatial distribution of hot spots on South beach
are shown in Figures 25 through 27 for WAL, SNO, and SSO. All beach segments contain hot
spot areas; however, the areas open during the nesting season to public recreation on WAL and
SNO had no hot spots (recreation areas are shown on Figures 25 and 26 with a purple boundary).
A high number of hot spots occurred within the vicinity of the Santa Ynez River mouth on the
southern end of WAL and northern end of SNO in all nesting categories. There are no hot spots
along the eastern edge of the beach segment. SSO hot spots occur to the south of Bear Creek in
every category along “border trail”.
52

Table 8. South Beach Hot Spot Categories
New Consecutive Sporadic
Initiated
WAL 4 17 18
SNO 7 15 38
SSO - 1 1
Hatched
WAL 1 4 17
SNO 1 13 16
SSO - - 1
Predated
WAL 4 9 6
SNO 9 12 24
SSO - 1 6

Figure 25. WAL - EHSA Results
53

Figure 26. SNO – ESHA Results

Figure 27. SSO - ESHA Results
54

The distribution of hots spots on WAL varied in all three nesting categories. Initiated
nests hot spots occur throughout WAL with many hexagons at the north end. Predated nesting
hot spots occur primarily on northern end of the segment, while hatched nests to the southern
end. SNO initiated and predated nesting hot spots have a similar spatial distribution. However,
hatched nest hot spots occur primarily on the northern and extreme southern end of the segment.
SSO had very few hots spots detected in only two hot spot categories. In each nesting
category there is a consistent hot spot in the vicinity of Bear Creek. Nest initiation has a
consecutive hot spot at the northern end while predated nest hot spots occur in two area above
Bear Creek.
5.3 Space-time Cube Hot Spot 3D Visualization
Visualizing the STC hot spot pattern showed only hot spots or non-statistically significant
trends in all beach segments. Figures 28 through 36 show each section for initiated nests while
predated and hatch nest categories are in Appendix D. On these figures, each bin column
contains 17 time-steps that represent each year of plover nests since 2002 with the most recent
year, 2018, at the top of the stack. Most stacks that were not statistically significant were
spatially distributed to the eastern edge of the habitat in all sections. The stacks that have no hot
spots remained consistent through all years. Those with hot spots remained in consistent areas
but varied in significance throughout the years.
MIN beach section did not show any hot spots in the ESHA, however, when looking at
all years in the STC visualization, the temporal distribution of hot spots can be observed. MIN
did not show any hot spots since 2014 but prior to that hot spots are shown predominately on the
northern end. Between 2009-2013 hot spots are seen on the sand sheets of northern MIN.
55

Figure 28. MIN - STC Initiated Nest Hot Spot Visualization.

Figure 29. SHN – STC Initiated Nest Hot Spot Visualization.
56

Figure 30. SHS – STC Initiated Nest Hot Spot Visualization.

Figure 31. North SAN – STC Initiated Nest Hot Spot Visualization.
57

Figure 32. South SAN – STC Initiated Nest Hot Spot Visualization.

Figure 33. WAL – STC Initiated Nest Hot Spot Visualization.
58

Figure 34. North SNO – STC Initiated Nest Hot Spot Visualization.

Figure 35. South SNO – STC Initiated Nest Hot Spot Visualization.
59

Figure 36. SSO – STC Initiated Nest Hot Spot Visualization.
SHN and SHS hot spots occur mainly on the beach front and foredune area earlier in the
time series except for central SHN sand sheet showing hot spots prior to 2013. Early in the time
series SHS had more hot spots at the northern end while later in the time series the southern end
showed an increase in hot spots.
SAN beach section contains the most bins containing hot spots within all stacks when
compared to all other beach sections (Figure 31 and 32). This supports the findings in the EHSA
showing SAN with most of the hot spots and in all categories. Hot spot distribution in the sand
sheets at south SAN end in 2013, like those on MIN and SHN. This suggests some anomaly,
when compared to all years, in 2013 that caused a shift in the hot spot distribution on north beach
further east on all segments.
On South beach, hot spots remain relatively consistent throughout, those stacks with no
hot spots remain that way through the all years while others show fluctuations. WAL hot spot
60

distribution remained on the southern end in most years but more recently heavily distributed on
the central and the northern areas, otherwise sporadic throughout. On SNO, hot spots were
distributed mainly at the northern end. In the EHSA, SNO open to recreation does not show any
hot spot categories. However, like MIN, the STC shows a few hot spots in early years, especially
at the south end (Figure 34 and 35). SSO also showed limited hot spots in the EHSA and in the
STC, most of the hot spots occur in recent years.
Hatched and predated STC visualizations show similar distributions, with several years
that appear to change distribution (Appendix D). Such as in 2010 and 2013 many predated hot
spots appear along SNO segment. Hatch hot spots on South beach seem to follow a similar
overall distribution as initiations.
5.4 Wrack Abundance
A total of 1713 nests were initiated between 2012-2018 on North and South beach of
VAFB. As described above, during these years, wrack accumulation was assessed weekly with
an abundance value given to each surveyed transect. During the data manipulation stage of this
research, each nest was assigned the wrack index value collected for the relevant transect during
the week when the nest was first established. All nests between 2012-2018 were included in this
analysis. This part of the analysis was designed to determine which wrack index value had the
highest percentage of initiated nests regardless of fate and of all successful hatched nests to see if
high wrack values correlate with higher percent of initiated nests or successfully hatched nests.
Surprisingly, a similar trend was found for North and South beach with a higher percentage of
nests in both categories having a low wrack index value of 1. Figure 37 summarizes these results
which are discussed in the following sections.
61

Figure 37. North and South Beach wrack values indicating the percent of initiated and percent
hatched nests per wrack index value for 2012-2018.
5.4.1 Initiated Nests Wrack Values
Approximately 50 percent of all South and 70 percent of all North beach nests were
initiated during weeks of level 1 wrack (low abundance). Level 2 wrack nests accounted for 30
percent all South and 26 percent all North nests respectively. Finally, level 3 and above wrack
nests accounted for 15 percent South and 7 percent North. There were very few initiated nests
during weeks of level 4 wrack values and none in level 5.
5.4.2 Hatched Nests Wrack Values
Successful clutch hatches made up 45 percent of all initiated nests (779) between 2012-
2018. Out of these nests 65 percent on North and 45 percent of South beach hatched nests
occurred during wrack index values at level 1. Approximately 30 percent of both North and
South successful hatched nests had level 2 wrack index with 6 percent of North and 15 percent of
0
20
40
60
80
1 2 3 4
% of Nests
North Beach Wrack Index
Initiated Percent
Hatched Percent
0
20
40
60
80
1 2 3 4 5
% of Nests
South Beach Wrack Index
Initiated Percent
Hatched Percent
62

South in level 3. Less than 5 percent of hatched nests on both beach sections had level 4 wrack
value with only 1 percent of hatched nest occurring in level 5 at South beach.

63

Chapter 6 Discussion and Conclusions
The plover population at VAFB has been monitored since it was listed as threatened by the
USFWS in 1992. However, only in the last 17 years have consistent data regarding nest success,
population fluctuation, and habitat been collected (Robinette et al. 2018; Mantech 2010).
Throughout the literature, habitat restoration and predator management have been identified as
the most prominent management strategy within the plover breeding range (USFWS 2007), each
site manger varying in approach and methodology. At VAFB, predator management successfully
occurs on the fly, without prior knowledge of possible predator hot spot trends. Understanding
where predator hot spots are will provide managers with preemptive primary focal areas. Large
scale restoration has been conducted with little prior research to guide project scope and
locations, often going about it in “good faith” that the plovers will use the area if invasive
vegetation is removed or if dunes are reconstructed. As a result, little is known about the impact
of restoration or whether these locations serve as hot spots for nesting plovers.
To understand the relationship between the distribution of nesting hot spots, predation,
and restoration 17 years of nesting data were used to assess spatial and temporal nesting. This
research addressed the following questions:
1) Are there consistent temporal and spatial nesting hot spots in three categories: nest
initiation, successful clutch hatch and clutch failure due to predation, and if so,
what are the key areas for focusing future predator management and restoration
work?
2) Do recent habitat restoration sites have hot spots of plover nesting pre-restoration
Have hot spot distributions changed post restoration in these areas?
3) Is there a correlation between high wrack abundance and nest initiation or hatch?
64

6.1 Result Overview
The data used in this study underwent an extensive overhaul in formatting and cleansing.
As a result, a streamlined file was created for use in the analyses and a conceptual model of a
relational database was structured for future studies. Three different forms of analysis were
conducted to determine the spatial distribution of nesting hot spots. Overall this study found that
hots spots do persist in certain areas of VAFB beaches annually and hot spot analysis can be
used as an effective tool in analyzing nesting patterns. Even though the annual hot spot analysis
did not provide useable results on the study site scale, it proved to be an excellent method to
assess restored beach areas pre- and post-restoration. STC and ESHA tools effectively captured
the spatial and temporal distribution of nesting at VAFB at the study site scale, providing
answers to the research questions in this thesis.
6.1.1 Are there consistent temporal and spatial nesting hot spots in three categories: nest
initiation, successful clutch hatch and clutch failure due to predation? What are the key
areas for focusing future predator management and restoration work?
While considering hot spots of nesting, the EHSA is the best tool to look at an overall
temporal view of the spatial nesting patterns. In this analysis, hot spot categories given more
consideration are those that are consecutive, persistent, and intensifying. In this discussion, these
categories are called significant hot spots. Hot spots classified as sporadic or new are not
considered significant for the purpose of this thesis since they occur in less than 90 percent of the
bins in a stack or, in the case of a new hot spot, never until the most recent year. The new hot
spot category could serve as an alert system if this analysis is run with each new year of data, but
here they are not considered for recommendation for restoration or predator activities for
broadscale management.
65

On North beach, the segment with the most significant hot spots in all nesting categories
is SAN followed by SHS (which is on the northern boundary of SAN). The most significant
hatch hot spots occur at the northernmost end surrounding the San Antonio River mouth.
Predated and initiated hot spots are scattered throughout SAN, however, significant predated hot
spots occur on the coastal areas throughout the northern half of the segment. Initiated nests share
a similar pattern as predated, however just south of the San Antonio river mouth, there is a large
hot spot.
A similar pattern is found on South beach, with the most significant hot spots occurring
north and south of the Santa Ynez River mouth. Both initiated and predated nest hot spots share a
similar pattern throughout, however most notably at the northern end near the area open for
public recreation.
This distribution supports the hypothesis that most initiated and hatched hot spots will
occur surrounding the river mouths. On South beach, predated hot spots on SNO appear to
concentrate near the areas open to recreation as well as the river mouths. This is possibly due to
the increase in human related activities, such as trash, drawing nest predators to the area. Areas
around the river mouths should continue to be areas of restoration consideration to improve
habitat quality. Predator management efforts should be focused along the river mouths as well as
areas adjacent to the public areas at WAL and SNO.
6.1.2 Which beaches of VAFB have hot spots of plover nesting and do these areas correlate
with recent habitat restoration activities? Have hot spot distributions changed post
restoration in these areas?
There are two management implications that result from locating hot spots of nest
initiation and nest hatching at VAFB. The first is that future restoration efforts can be focused in
hot spot areas with the assumption that habitat enhancement will improve and provide better
66

habitat quality. The second is to assess past restoration efforts to determine if these efforts took
place in a hot spot area. MIN and SHN beach segments have had relatively no hot spots in five
years according to exploring the STC visualization and overall no significant hot spots were
found in the EHSA; and yet restoration has ensued in this area within the last four years.
Although the effects of this restoration on the plover breeding population at VAFB may have yet
to be expressed, no significant increase in nesting has occurred there in the last 4 years since
restoration (Robinette et al. 2018). Nelson (2007) points out that without plover social cues or
high numbers of actively nesting plovers in the area, restoration in areas where there are no hot
spots of activity may fail to bring an increase in prospective breeders. Therefore, the results of
“good faith” restoration may prove difficult in increasing nesting (Ahlering and Faaborg 2006).
Exploring the possible reasons MIN through SHS lack high number of breeding plovers
compared to other beaches (in spite of its available habitat) would be important in attempting to
explain the southern population shift at North beach.
Northern SNO and all of WAL beach also underwent a large restoration, however this
area had a significant amount of hot spot locations near the restoration area prior to restoration.
Similar to Zarnestke (2010) findings in the Pacific Northwest Coast, at VAFB there was first a
substantial increase in nesting in the two years following dune contouring followed by a drop off
in nesting activity as the dunes began to fill in with native vegetation and dune slopes began to
be reshaped naturally by beach processes. At SNO, new hot spots of nesting and hatching have
been relatively consistent post-restoration (2014-2017), however, in 2018 no hot spots were
observed. Tracking this pattern into the future will determine if restoration efforts will follow a
similar declining fate without follow-up dune contouring (Zarnestke 2010). WAL beach had
nesting and hatched hot spots prior to restoration in that area and although the distribution did
67

not change significantly after restoration, hot spots size did increase during years that hot spots
were present on the edges of the restoration boundary.
While comparing hot spots pre- and post-restoration does not give an overall analysis on
plover productivity, it is a good indicator of important areas of nesting that are statistically
significant. Powell and Collier (2000) recommend yearly evaluation of these restored nesting
areas. This evaluation should not be based on the presence of nesting alone but tracking the
productivity. Therefore, management should continue to track the response to restoration by the
plovers annually. HSA is a powerful tool to detect changes in hot spot distribution and offers
insight on the important nesting areas.
6.1.3 Is there a correlation between high wrack abundance and nest initiation or hatch?
This section of the research was an exploratory study. It is possible that food availability
is not linked to nest site location because adult plovers are able to move up and down beaches for
feeding. However, Colwell (2007) found that plover chicks less than three days old have
significantly higher mortality rates than older chicks. When chicks hatch, they need to feed and
traveling long distances for food or when energy intake is low, can cause slow growth in chicks
which can increase mortality. In addition, traveling in search of food combined with low wrack
abundance keeps chicks exposed for longer periods of time to predation risks. Exploring chick
survival and food resources at VAFB would be a worthwhile research endeavor.
Though the results to this simple wrack exploration did not support the original
hypothesis, that more nests would be established in areas with higher wrack abundance, this is a
good step towards looking at how ocean subsidies may influence nesting. Many papers describe
the correlation between high shorebird food abundance and high wrack deposits on beaches in
the wintering season (Dugan 2003; Lafferty 2006), however, there is no literature regarding food
68

abundance during the breeding season of plovers. While this thesis focused on restoration and
predator management, other factors such as food availability may have a great impact on nest
distribution at VAFB and should be explored.
6.2 Management Recommendations and Future Work
This study of spatial and temporal distributions of nesting hot spots and the relationship
with management activities can be explored further in many ways. This thesis built a framework
in which more questions can be studied, and additional exploration of the data can be initiated.
Continuing yearly analysis of each breeding season will be valuable in understanding nesting
trends in the future. While, plovers are one of the important threatened species on VAFB,
applying this method of analysis to other species that share the beach ecosystem with the plover
would help when applying management activities such as restoration to improve habitat for
multiple species and understanding the impact of contouring or herbicide to non-target species.
This analysis lumped all predated nests into one category that included both coyote and
raven predation. Predator control strategies varies significantly due to the terrestrial versus aerial
nature of these two species. Nest predation hot spots should be evaluated in the future with these
two categories to determine hot spots for each species and if there are any differences in where
these two predators are taking nests. In addition, future EHSA should consider updating the
polygon mask used in the parameter settings. Most of the stacks in the STC had no data resulting
in no significant ranking in all HSA hexagons. This may impact the neighborhood statistical
analysis in these tools. Consideration should also be given to reducing the amount of years used
in the EHSA to determine if trends change using only most recent years. The tool requires at
least six time-steps or six years of plover nesting data.
69

The focus in this study was on nesting distribution, however, the success of plovers is not
merely the successful hatching of eggs, it is the survival and fledging of these chicks. Further
study should include spatial analysis of where fledge rates are the highest and if there are any hot
spots in fledging. This can be achieved by sorting and adding the data field of whether a nest had
a successful fledge and running the ESHA on those nests. However, one issue with this
methodology would be that some broods or chicks move in and out of their territory during the
chick rearing stage. Therefore, not being tied to the nesting location habitat factors contributing
to the successful fledge would need to be addressed.
6.3 Conclusions
While hot spot analysis in ArcGIS has been used widely in the health care, policing,
municipal and utilities industries it has only recently been used in conservation management to
identify “hotspots” in environmental data whether it be species, population, or roadside wildlife
mortality (Kazemi et al. 2016; McLemore 2017). This study has shown the value of using this
tool to determine nesting hot spots and is a good template for other plover breeding sites or
analysis on nesting distributions in general. Hot spots in nesting are prevalent at VAFB and
should be analyzed yearly to guide management activities and restoration.

70

References
Ahlering, Marissa A. and John Faaborg, 2006. “Avian Habitat management Meets Conspecific
Attraction: If You Build It, Will They Come?” The Auk 123, no. 2: 301-312.
Birch, Colin, Oom Sander, and Jonathan Beecham. 2007. “Rectangular and Hexagonal Grids
Used for Observation, Experiment, and Simulation in Ecology.” Ecological Modeling
206, no. 3-4: 347-359.
Boarman, W.I. and B. Heinrich. 1999. “Common Raven (Corvus corax).” The Birds of North
America no. 476.
Brindock, Kevin M., and Mark A. Colwell. 2011. "Habitat Selection by Western Snowy Plovers
during the Nonbreeding Season." Journal of Wildlife Management 75, no. 4: 786-793.
Camp, R.J. and R.L. Knight. 1993. “Common Raven populations in Joshua Tree National
Monument, California.” Western Birds 24: 198-199.
Colwell, M. A., C. B. Millett, J. J. Meyer, J. N. Hall, S. J. Hurley, S. E. McAllister, A. N.
Transou, and R. R. LeValley. 2005. "Snowy Plover Reproductive Success in Beach and
River Habitats." Journal of Field Ornithology 76, no. 4: 373-382.
Dugan, Jennifer E., David M. Hubbard, Michael D. McCrary, and Mark O. Pierson. 2003. “The
Response of Macrofauna Communities and Shorebirds to Macrophyte Wrack Subsidies
on Exposed Sandy Beaches of Southern California.” Estuarine, Coastal and Shelf
Science no. 58: 25-40.
Eberhart-Phillips, Luke J., Brian R. Hudgens, and Mark A. Colwell. 2016. “Spatial Synchrony of
a Threatened Shorebird: Regional Roles of Climate, Dispersal and Management." Bird
Conservation International 26, no. 1: 119-135.
Esri. 2018a. “How Create Space Time Cube By Aggregating Points Works.” Accessed on April
20, 2019. https://pro.arcgis.com/en/pro-app/tool-reference/space-time-pattern-
mining/create-space-time-cube.htm.
Esri. 2018b. “How Emerging Hot Spot Analysis Works.” Accessed on April 20, 2019.
http://desktop.arcgis.com/en/arcmap/10.3/tools/space-time-pattern-mining-
toolbox/emerginghotspots.htm.
Fahy, Krista, Rothstein, Steve I., Stewart-Oaten, Allan, and Lafferty, Kevin. 2008. “The effect of
habitat choice, density and distribution on breeding western snowy plovers.” ProQuest
Dissertations Publishing, n.d.
Getis. A. and J.K. Ord. 1992. “The Analysis of Spatial Association by Use of Distance Statistic.”
Geographical Analysis 24, no. 3: 189-206.
71

Gonçalves, Sílvia C. and João C. Marques. 2011. "The Effects of Season and Wrack Subsidy on
the Community Functioning of Exposed Sandy Beaches." Estuarine, Coastal and Shelf
Science 95, no.1: 165-177.
Hardy, Michael A., and Mark A. Colwell. 2012. "Factors Influencing Snowy Plover Nest
Survival on Ocean-Fronting Beaches in Coastal Northern California." Waterbirds 35, no.
4: 503-656.
Hubbard, David M. and Jenifer E. Dugan. 2003. "Shorebird use of an Exposed Sandy Beach in
Southern California." Estuarine, Coastal and Shelf Science 58: 41-54.
Kandel, Sean, Jeffrey heer, Catherin Plaisant, Jessie Kennedy, Frank van Ham, Nathalie Henry
Riche, Chris Weaver, Bongshin Lee, Dominque Brodbeck, and Paolo Buono. 2001.
“Research directions in data wrangling: Visualizations and transformations for usable and
credible data.” Information Visualization 104, no. 4: 271-288.
Kazemi, Vahed Dehghani, Hamidreza Jafari, and Ahmadreza Yavari. 2016. "Spatio-Temporal
Patterns of Wildlife Road Mortality in Golestan National Park-North East of Iran." Open
Journal of Ecology 6, no. 6: 312-324.
Kendall. M. and J.D. Gibbons. 1990. “Rank Correlation Methods.” Oxford University Press,
Oxford.
Kristan, W.B. and Boarman. 2007. Effects of anthropogenic developments on Common Raven
nesting biology in the west Mojave Desert, California. Ecological Applications 17, no. 6:
1703-1713.
Ince, Rebecca, Glenn A. Hyndes, Paul S. Lavery, and Mathew A. Vanderklift. 2007. “Marine
Macrophytes Directly Enhance Abundances of Sandy Beach Fauna Through Provision of
Food and Habitat.” Estuarine Coastal and Shelf Science 74: 77-86.
Lafferty, Kevin, Darcie Goodman, and Cristina Sandoval. 2006. "Restoration of Breeding by
Snowy Plovers Following Protection from Disturbance." Biodiversity and Conservation
15, no. 7: 2217-2230.
Lee, Mong L., Tok Wang Ling, and Wai Lup Low. 2000. “A Knowledge-based Intelligent Data
Cleanser.” In Proc. Of the Sixth ACM SIGKDD Int. Conf. on Knowledge Discovery and
Data Mining, pages 290-294.
Leja, Stephanie. “Habitat Selection and Response to Restoration by Breeding Western Snowy
Plovers in Coastal Northern California.” 2015. Master’s Thesis, Humboldt State
University.
Mann, H.B. 1945. “Nonparametric Tests Against Trends.” Econometrica 13, no. 3: 245-259.
ManTech SRS Technologies, Inc. 2018. Western Snowy Plover Habitat Restoration at
Vandenberg Air Force Base, California: Minuteman Dune Restoration – Year 3
Summary.
72

McLemore, Shannon. 2017. “Spatio-Temporal Analysis of Wildfire Incidence in the State of
Florida.” Master’s Thesis, University of Southern California.
Montello, Daniel R., and Paul C. Sutton. 2006. An Introduction to Scientific Research Methods
in Geography & Environmental Studies. Thousand Oaks: SAGE Publications, Inc.
Muir, Jordan J., and Mark A. Colwell. 2010. "Snowy Plovers Select Open Habitats for Courtship
Scrapes and Nests." The Condor 112, no. 3: 507-510.
Muller H and JC Freytag. 2005. “Problems, methods and challenges in comprehensive data
cleansing.” Technical Report HUB-IB_164, ed. Humboldt-Universitat.
Neuman, Kristen K. 2004. “Effect of Mammalian Predator Management on Snowy Plover
Breeding Success.” Waterbirds 27, no. 3: 257-263.
Page, Gary W., and Lynn E. Stenzel (eds.). 1981. “The breeding status of the snowy plover in
California.” Western Birds 12, no. 1: 1-40.
Page, Gary W., Lynne E. Stenzel, David W. Winkler, and Christopher W. Swarth. 1983.
"Spacing Out at Mono Lake: Breeding Success, Nest Density, and Predation in the
Snowy Plover." The Auk 100 no. 1: 13-24.
Page, Gary W., Lynne E. Stenzel, and Christine A. Ribic. 1985a. "Nest Site Selection and Clutch
Predation in the Snowy Plover." The Auk 102, no. 2: 347-353. doi:10.2307/4086778.
Page, Gary W., Stenzel, Lynne E., Schuford, David W., and Charles R. Bruce. 1991.
“Distribution and abundance of the snowy plover on its western North American
breeding grounds.” Journal of Field Ornithology 62:245-255.
Palacios, E., L. Alfaro, and Gary W. Page. 1994. “Distribution and abundance of breeding snowy
plovers on the Pacific Coast of Baja, California.” Journal of Field Ornithology 65:490-
497.
Patrick, Allison M. K. and Mark A. Colwell. 2018. "Annual Variation in Distance to Nearest
Neighbor Nest Decreases with Population Size in Snowy Plovers." Wader Study 124
no. 3.
Powell, Abbey N. 2001. "Habitat Characteristics and Nest Success of Snowy Plovers Associated
with California Least Tern Colonies." The Condor 103, no. 4: 785-792.
Rahm, Erhard and Hong Hai Do. 2000. “Data Cleaning: Problems and Current Approaches.”
IEEE Data Engineering Bulletin 23, no. 4.
Robinette, D.P, J.K. Miller, and J. Howar. 2017. “Monitoring and Management of the
Endangered California Least Tern and the Threatened Western Snowy Plover at
Vandenberg Air Force Base, 2017.” Unpublished Report, Point Blue Conservation
Science, Petaluma, CA. Point Blue Contribution No. 2154.
73

Robinette, D.P, J.K. Miller, and J. Howar. 2018. “Monitoring and Management of the
Endangered California Least Tern and the Threatened Western Snowy Plover at
Vandenberg Air Force Base, 2019.” Unpublished Report, Point Blue Conservation
Science, Petaluma, CA.
Saalfeld, Sarah T., Warren C. Conway, David A. Haukos, and William P. Johnson. 2012.
"Snowy Plover Nest Site Selection, Spatial Patterning, and Temperatures in the Southern
High Plains of Texas." The Journal of Wildlife Management 76, no. 8: 1703-1711.
Schlacher, Thomas A., Briony M. Hutton, Ben L. Gilby, Nicholaus Porch, Grainne S. Maguire,
Brooke Maslo, Rod M. Connolly, Andrew D. Olds, and Michael A. Weston. 2017. "Algal
Subsidies Enhance Invertebrate Prey for Threatened Shorebirds: A Novel Conservation
Tool on Ocean Beaches?" Estuarine, Coastal and Shelf Science 191: 28-38.
SRS Technologies, Inc. 2005. Final Plan for the Removal of Selected Invasive Plants from
Western Snowy Plover Habitat at Vandenberg Air Force Base.
USFWS. 2007. Recovery Plan for the Pacific Coast Population of the Western Snowy Plover
(Charadrius alexandrinus nivosus). California/Nevada Operations Office, Sacramento,
California. 274pp plus Appendices.
Warriner, John S., Jane C. Warriner, Gary W. Page, and Lynne E. Stenzel. 1986. "Mating
System and Reproductive Success of a Small Population of Polygamous Snowy Plovers."
The Wilson Bulletin 98, no. 1: 15-37.
Zarnetske, Phoebe L., Eric W. Seabloom, and Sally D. Hacker. 2010. "Non-Target Effects of
Invasive Species Management: Beachgrass, Birds, and Bulldozers in Coastal Dunes."
Ecosphere 1, no. 5: 20.

74

Appendix A. Extract of Plover Nesting Data Post Cleaning
Nest ID N UTM84 E UTM84 Beach
Section
Monitoring
Segment
Transect Block Initiation
Date
Date Found Wrack
Index at
Initiation
EWF Week
Num
Clutch
Size
Clutch
Completion
Fate Date Fate Pred/
Cause
Days
Active
15SAN034 3851276 717205 North San Antonio TURTLE 4/27/2015 4/30/2015 0.5 2 18 3 5/1/2015 5/29/2015 Hatched

31
15SAN043 3850853 716925 North San Antonio NARROWS 4/26/2015 5/14/2015 0.5 3 17 3

5/27/2015 Hatched

31
15SAN061 3852643 717564 North San Antonio DEADBUG 5/25/2015 5/26/2015 0.5 1 22 2 5/27/2015 6/26/2015 Hatched

32
15SAN063 3851575 717183 North San Antonio WISHBONE 5/20/2015 5/26/2015 0.5 3 21 3

6/20/2015 Hatched

32
15SAN069 3851858 717270 North San Antonio TRASH 5/12/2015 6/1/2015 0.5 3 20 3

6/12/2015 Hatched

31
15SAN080 3851711 717227 North San Antonio PICKUPSTIX 6/10/2015 6/10/2015 0.5 1 24 3 6/15/2015 7/12/2015 Hatched

31
15SAN083 3851532 717224 North San Antonio WISHBONE 6/11/2015 6/12/2015 0.5 1 24 3 6/15/2015 7/13/2015 Hatched

28
15SAN085 3851749 717362 North San Antonio PICKUPSTIX 6/14/2015 6/15/2015 0.5 1 24 3 6/18/2015 7/15/2015 Hatched

31
15SAN088 3850795 716916 North San Antonio NARROWS 6/11/2015 6/17/2015 0.5 2 24 2

7/12/2015 Hatched

31
15SAN089 3850933 716950 North San Antonio NARROWS 6/14/2015 6/17/2015 0.5 2 24 3 6/18/2015 7/12/2015 Hatched

31
15SAN091 3851172 717052 North San Antonio TURTLE 6/14/2015 6/17/2015 0.5 2 24 3 6/18/2015 7/15/2015 Hatched

31
15SAN093 3851483 717166 North San Antonio GROSS 6/3/2015 6/17/2015 0.5 3 23 3

7/4/2015 Hatched

31
15SAN094 3851943 717295 North San Antonio JUGHEAD 6/6/2015 6/17/2015 0.5 3 23 3

7/7/2015 Hatched

31
15SAN095 3852472 717465 North San Antonio TEETER 6/12/2015 6/17/2015 0.5 3 24 3

7/13/2015 Hatched

31
15SAN103 3853556 717687 North San Antonio SHSEND 6/13/2015 6/26/2015 0.5 3 24 3

7/14/2015 Hatched

31
15SAN105 3853330 717636 North San Antonio L13BUOY 5/30/2015 6/26/2015 0.5 2 22 2

6/30/2015 Hatched

32
15SAN109 3852646 717624 North San Antonio DEADBUG 5/29/2015 6/29/2015 0.5 3 22 3

6/29/2015 Hatched

31
15SAN124 3851930 717290 North San Antonio TRASH 6/8/2015 8/6/2015 0.5 0 24 3

7/9/2015 Hatched

32
15SHN002 3856151 718188 North Shuman North ROLLINGROCK 4/5/2015 4/13/2015 0.5 2 14 2 4/7/2015 5/7/2015 Hatched

31
15SHS005 3855434 718063 North Shuman South BUOYFARM 4/24/2015 5/5/2015 0.5 3 17 3

5/25/2015 Hatched

35
15SHS007 3853797 717774 North Shuman South TRILOG 5/13/2015 5/14/2015 0.5 1 20 3 5/17/2015 6/14/2015 Hatched

32
15SHS012 3854843 717946 North Shuman South OJ 6/12/2015 6/22/2015 0.5 3 24 3

7/13/2015 Hatched

36
15SNO014 3838789 718871 South Surf North DRIFTWOOD 4/9/2015 4/9/2015 0.5 1 15 3 4/13/2015 5/14/2015 Hatched

28
15SSO015 3838228 718683 South Surf South SQUID 4/28/2015 4/28/2015 0.5 1 18 3 5/2/2015 5/30/2015 Hatched

5
15WAL004 3843262 719708 South Wall WALL OPEN 4/3/2015 4/3/2015 0.5 1 14 3 4/8/2015 5/9/2015 Hatched

19
15SAN003 3853364 717672 North San Antonio L13BUOY 3/27/2015 3/30/2015 0.5 2 13 3 3/31/2015 4/24/2015 Predated Unknown 16
15SAN068 3853333 717656 North San Antonio L13BUOY 5/28/2015 5/29/2015 0.5 1 22 1

6/2/2015 Predated Unknown 25
15SAN102 3853564 717701 North San Antonio SHSEND 6/22/2015 6/26/2015 0.5 3 26 3

7/11/2015 Predated Unknown 9
15SAN104 3853502 717676 North San Antonio ARROWHEAD 6/25/2015 6/26/2015 0.5 1 26 2 6/27/2015 7/11/2015 Predated Unknown 4
15SHS010 3855072 717993 North Shuman South MINIDUMP 6/9/2015 6/10/2015 0.5 1 24 3 6/23/2015 7/4/2015 Predated Unknown 4
15SAN010 3852213 717956 North San Antonio SHORTRIBS 4/7/2015 4/8/2015 0.5 1 15 3 4/11/2015 4/16/2015 Predated Skunk 4
15SSO010 3838561 718774 South Surf South METEOR 4/12/2015 4/13/2015 0.5 1 15 1

4/16/2015 Failed Tide 5
15SHN001 3856180 718191 North Shuman North ROLLINGROCK 4/2/2015 4/3/2015 0.5 1 14 1

4/6/2015 Failed Wind 2

75

Appendix B. Emerging Hot Spot Analysis Categories
(Source: Esri 2018b)
Pattern Name Definition
No Pattern Detected There were not hot or cold spot patterns found in area
New Hot Spot A location that has never been a hot spot before but is a hot spot for the final
time step
Consecutive Hot Spot A location with an uninterrupted run of hot spot bins in the final time-step
intervals. Less than 90 percent of all bins are hot spots and the location has never
been a hot spot prior to the final run.
Intensifying Hot Spot A location that has been a hot spot for 90 percent of the time-step intervals
including the final interval. The intensity of clustering of high counts in each
time step is increasing overall and the increase is statistically significant.
Persistent Hot Spot A location that has been a hot spot for 90 percent of the time-step intervals with
no apparent trend indicating an increase or decrease in the intensity of clustering
over time.
Diminishing Hot Spot A location that has been a hot spot 90 percent of the time-step intervals including
the final interval. The intensity of the clustering in each step is decreasing overall
with the decrease being statistically significant.
Sporadic Hot Spot Areas that are hots spots at one time-step interval then not a hot spot.
Less than 90 percent of the time-step intervals have been hot spots and never
have they been cold spots.
Oscillating Hot Spot A hot spot for the final time-step interval that has a history of being a cold spot
during the prior interval. Less than 90 percent of the intervals have been
statistically significant hot spots.
Historical Hot Spot At least ninety percent of the time-step intervals have been hot spots, but the
most recent time period is not hot.
New Cold Spot A location that has never been a hot spot before but is a hot spot for the final
time step
Consecutive Cold Spot A location with an uninterrupted run of hot spot bins in the final time-step
intervals. Less than 90 percent of all bins are hot spots and the location has never
been a hot spot prior to the final run.
Intensifying Cold Spot A location that has been a hot spot for 90 percent of the time-step intervals
including the final interval. The intensity of clustering of high counts in each
time step is increasing overall and the increase is statistically significant.
Persistent Cold Spot A location that has been a hot spot for 90 percent of the time-step intervals with
no apparent trend indicating an increase or decrease in the intensity of clustering
over time.
Diminishing Cold Spot A location that has been a hot spot 90 percent of the time-step intervals including
the final interval. The intensity of the clustering in each step is decreasing overall
with the decrease being statistically significant.
Sporadic Cold Spot Areas that are hots spots at one time-step interval then not a hot spot.
Less than 90 percent of the time-step intervals have been hot spots and never
have they been cold spots.
Oscillating Cold Spot A hot spot for the final time-step interval that has a history of being a cold spot
during the prior interval. Less than 90 percent of the intervals have been
statistically significant hot spots.
Historical Cold Spot At least ninety percent of the time-step intervals have been hot spots, but the
most recent time period is not hot.
76

Appendix C. WAL and SNO Restoration Areas Hatched HSA

Figure 38. WAL and SNO Restoration Areas Hatched HSA 2004-2006.

Figure 39. WAL and SNO Restoration Areas Hatched HSA 2007-2009.
77

Figure 40. WAL and SNO Restoration Areas Hatched HSA 2010-2012.

Figure 41. WAL and SNO Restoration Areas Hatched HSA 2013-2015.
78

Figure 42. WAL and SNO Restoration Areas Hatched HSA 2016-2018.

79

Appendix D. STC Hot Spot Visualization Hatched and Predated

Figure 43. MIN – STC Hatched Nest Hot Spot Visualization.

Figure 44. MIN – STC Predated Nest Hot Spot Visualization.
80

Figure 45. SHN – STC Hatched Nest Hot Spot Visualization.

Figure 46. SHN – STC Predated Nest Hot Spot Visualization.
81

Figure 47. SHS – STC Hatched Nest Hot Spot Visualization.

Figure 48. SHS – STC Predated Nest Hot Spot Visualization.
82

Figure 49. North SAN – STC Hatched Nest Hot Spot Visualization.

Figure 50. North SAN – STC Predated Nest Hot Spot Visualization.
83

Figure 51. South SAN – STC Hatched Nest Hot Spot Visualization.

Figure 52. South SAN – STC Predated Nest Hot Spot Visualization.
84

Figure 53. WAL – STC Hatched Nest Hot Spot Visualization.

Figure 54. WAL – STC Predated Nest Hot Spot Visualization.
85

Figure 55. North SNO- STC Hatched Nest Hot Spot Visualization.

Figure 56. North SNO – STC Predate Nest Hot Spot Visualization.
86

Figure 57. South SNO – STC Hatched Nest Hot Spot Visualization.

Figure 58. South SNO – STC Predated Nest Hot Spot Visualization.
87

Figure 59. SSO – STC Hatched Nest Hot Spot Visualization.

Figure 60. SSO – STC Predated Nest Hot Spot Visualization.

Abstract (if available)

Abstract The population decline of Western snowy plover (Charadrius nivosus nivosus) and subsequent listing as a threatened species by the U.S. Fish and Wildlife Service (USFWS) along the Pacific Coast, is a result of poor reproductive success that is considered directly related to habitat loss and nest predation. Habitat restoration and predator management are active key components to the recovery plan of this species. Usually “good faith” restoration plans are carried out often without the site-specific understanding of nest distribution and other factors that influence nesting to focus these efforts. Bottom-up factors, such as food availability, may contribute to the nest initiation by courting adult Western snowy plovers but these factors have not been directly assessed during the breeding season. Habitat varies between breeding sites

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Evaluating predator prey dynamics and site utilization patterns of golden eagles using resource selection modeling and spatiotemporal pattern mining

PDF

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

PDF

Building better species distribution models with machine learning: assessing the role of covariate scale and tuning in Maxent models

PDF

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

PDF

Spatial analysis of human activities and wildfires in the Willamette National Forest

PDF

A comparison of GLM, GAM, and GWR modeling of fish distribution and abundance in Lake Ontario

PDF

Preparing for immigration reform: a spatial analysis of unauthorized immigrants

PDF

A spatial and temporal exploration of how satellite communication devices impact mountain search and rescue missions in California’s Sierra Nevada mountain range

PDF

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

PDF

Quantifying changes in monarch butterfly habitat in California

PDF

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

PDF

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

PDF

Congestion effects on arterials as a result of incidents on nearby freeway: When should you get off the highway?

PDF

Increase in surface temperature and deep layer nitrate in the California Current: a spatiotemporal analysis of four-dimensional hydrographic data

PDF

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

PDF

A spatio-temporal analysis of the metabolizable energy intake of the Tana River red colobus (Procolobus rufomitratus rufomitratus)

PDF

Use of GIS for analysis of community health worker patient registries from Chongwe district, a rural low-resource setting, in Lusaka Province, Zambia

PDF

A Maxent-based model for identifying local-scale tree species richness patch boundaries in the Lake Tahoe Basin of California and Nevada

PDF

Species distribution modeling to predict the spread of Spartium junceum in the Angeles National Forest

PDF

Using VHSR multispectral imagery and object-based extraction to discover vernal meadows through vegetative persistence at Fort Ord, California

Asset Metadata

Creator Butala, Regina Mary (author)

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

School College of Letters, Arts and Sciences

Degree Master of Science

Degree Program Geographic Information Science and Technology

Publication Date 11/12/2019

Defense Date 11/11/2019

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

Tag beach ecosystem,hot spot analysis,nest density,nest distribution,OAI-PMH Harvest,plover,shorebird nesting,snowy plover,space time cube,spatial analysis

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Kemp, Karen K (committee chair), Loyola, Laura Cyra (committee member), Ruddell, Darren Martin (committee member)

Creator Email raven4sage@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-232802

Unique identifier UC11674729

Identifier etd-ButalaRegi-7911.pdf (filename),usctheses-c89-232802 (legacy record id)

Legacy Identifier etd-ButalaRegi-7911.pdf

Dmrecord 232802

Document Type Thesis

Rights Butala, Regina Mary

Type texts

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

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

Repository Name University of Southern California Digital Library

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