The Evergreen State College Institutional Repository

The effects of temperature and humidity on Taylor’s Checkerspot butterfly life stage length and development

Item

Title

The effects of temperature and humidity on Taylor’s Checkerspot butterfly life stage length and development

Creator

Dunning, Ray

Identifier

Thesis_MES_2023_DunningR

extracted text

EFFECTS OF TEMPERATURE AND HUMIDITY ON TAYLOR’S CHECKERSPOT
BUTTERFLY LIFE STAGE LENGTH AND DEVELOPLMENT: RESEARCH IN
COLLABORATION WITH THE SUSTAINABILITY IN PRISONS PROJECT AND
INCARCERATED TECHNICIANS

By
R.E. Dunning

A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June 2023

©2023 by Raychel Dunning. All rights reserved.

This Thesis for the Master of Environmental Studies Degree
by
R.E. Dunning

has been approved for
The Evergreen State College
by

___________________________________
John Withey, Ph. D.
Member of the Faculty

June 16, 2023
Date

ABSTRACT:
Effects of Temperature and Humidity on Taylor’s Checkerspot Butterfly Life Stage Length and
Development: Research in Collaboration with the Sustainability in Prisons Project and
Incarcerated Technicians

Raychel E. Dunning

Captive rearing programs are used in conservation research to prevent critically endangered
species from becoming extinct and increasing their populations. One additional benefit of captive
rearing programs is the ability to collect data and conduct research and analysis on endangered
species reared in captivity within a more controlled environment. Environmental conditions are
especially important to understand for species’ whose internal processes are regulated by
external conditions, such as butterflies. The Taylor’s checkerspot butterfly (Euphydryas editha
taylori) is a federally endangered butterfly found in imperiled prairie areas of the Puget
Lowlands, Oregon’s Willamette Valley, and parts of British Columbia, Canada. In 2012 the
Sustainability in Prisons Project’s butterfly program was established at Mission Creek
Correctional Center for Women (MCCCW). This captive rearing program relies on keeping
strictly controlled environmental conditions and meeting temperature and humidity targets to
successfully raise the butterfly through its life stages. Conditions in the wild are gradually
experiencing hotter summers and warmer, wetter winters under climate change. This may result
in additional risks to the butterfly’s survival, such as resulting in a phenological mismatch with
its host species, Plantago lanceolata. In order to determine to what extent increases in
temperature and humidity influence the Taylor’s checkerspot life stage length and development,
an analysis of average temperature and average relative humidity was conducted. Environmental
conditions for the 2021 and 2022 years at MCCCW were compared to the length of different
instar stages, and additional growing degree day calculations were completed to obtain an idea as
to if GDD could be a suitable metric to use in future studies. The influence of average
temperature and average relative humidity on different instar stages and periods of development
were variable between the two years and greenhouses, but increased temperature typically
decreased the time spent within a given life stage, while relative humidity was shown to weakly
increase time spent in an instar stage.

TABLE OF CONTENTS
ABSTRACT: ............................................................................................................................................... 3
LIST OF FIGURES ................................................................................................................................... vi
LIST OF TABLES ................................................................................................................................... viii
ACKNOWLEDGEMENTS ...................................................................................................................... ix
ACRONYMS .............................................................................................................................................. xi
Introduction ................................................................................................................................................. 1
Literature Review ....................................................................................................................................... 5
Growing Degree Days (GDD) – General uses ......................................................................................... 6
Taylor’s Checkerspot Butterfly – General Overview ................................................................................ 7
Euphydryas editha ................................................................................................................................ 7
Larval development and behavior......................................................................................................... 7
History and Endangerment Status ............................................................................................................ 9
Phenological asynchrony and mismatch................................................................................................. 11
Phenological changes in butterflies under climate change .................................................................... 13
Current Conservation Measures ............................................................................................................. 14
Institute for Applied Ecology and Oregon Zoo efforts ........................................................................ 14
Sustainability in Prisons Project and MCCCW .................................................................................. 16
Coffee Creek Correctional Facility and Oregon Fish & Wildlife ....................................................... 18
Conclusion .............................................................................................................................................. 21
Methods...................................................................................................................................................... 22
Site .......................................................................................................................................................... 23
Environmental Data ................................................................................................................................ 26
Collection of Wild Females ..................................................................................................................... 28
Procedures and life stages ...................................................................................................................... 30
Oviposition Experimental Setup and Daily Care ................................................................................ 30
Egg to third instar daily care and experimental set-up ...................................................................... 33
Third Instar to Diapause ..................................................................................................................... 35
Diapause ............................................................................................................................................. 37
Data Analysis .......................................................................................................................................... 39
Life stage length and development ...................................................................................................... 39

iv

Environmental Data ............................................................................................................................ 41
Growing Degree Days ........................................................................................................................ 42
JMP ..................................................................................................................................................... 42
Results ........................................................................................................................................................ 43
Collected to hatch (egg stage) ................................................................................................................ 43
Hatch to 2nd Instar (1st instar stage) ..................................................................................................... 44
2nd to 3rd Instar (2nd instar stage) ............................................................................................................ 46
Hatch to 3rd Instar ................................................................................................................................... 48
Collected to Third Instar ......................................................................................................................... 50
Third to Fifth Instar Life Stage ............................................................................................................... 51
Collected to 5th instar (diapause): .......................................................................................................... 53
Discussion .................................................................................................................................................. 56
Limitations .............................................................................................................................................. 60
Future research....................................................................................................................................... 61
References .................................................................................................................................................. 63

v

LIST OF FIGURES
Figure 1. Evergreen State College graduate students and SPP coordinators Raychel Dunning
and Jen Bass assisting butterfly technicians in butterfly husbandry care. ................................... 22
Figure 2. Mission Creek Correctional Center for Women’s Taylor’s checkerspot butterfly
program area. ............................................................................................................................... 24
Figure 3. Image of greenhouses used for Taylor’s checkerspot butterfly captive-rearing and
breeding. Left is Raven greenhouse; right is Turtle greenhouse. Plantain beds near the
greenhouses are covered in cloches during the colder, winter months. ....................................... 25
Figure 4. The two greenhouses and the shed are located directly next to the prison’s perimeter
fence. ............................................................................................................................................. 26
Figure 5. Wild gravid females are collected in glass containers and placed within coolers for
transportation from the field to the prison.................................................................................... 29
Figure 6. Oviposition chamber with honey dome and water sponge. Image by the Sustainability
in Prisons Project. ........................................................................................................................ 32
Figure 7. Oviposition chamber enclosure with a honey dome and soaked sponge to nourish an
adult Taylor’s checkerspot butterfly. Two HOBO loggers per greenhouse rack are placed in
replicate butterfly-free chambers. Image by the Sustainability in Prisons Project. ..................... 32
Figure 8. Image by SPP, from SPP’s Eggs to Third Instar procedures. Eggs are cut from leaf
and placed into a 5.5oz cup, where they are then stored inside plastic shoe bins with a saturated
blue paper towel. ........................................................................................................................... 33
Figure 9. Photo by SPP, as found in their Eggs to Third Instar procedures................................ 34
Figure 10. Example of the 16oz cup label created when larvae are transferred from 5.5oz egg
cups into 16oz third instar cups. Image credit Sustainability in Prisons Project. ....................... 35
Figure 11. Image from SPP Diapause procedures demonstrating which larvae are considered
“moved” during movement checks. Image credit: Sustainability in Prisons Project. ................. 38
Figure 12. HOBO loggers are placed under terracotta pots in the outside shed during diapause
to mimic the conditions of the larvae. Image by the Sustainability in Prisons Project. ............... 38
Figure 13. Larvae in diapause are placed under terracotta pots. Image credit to the
Sustainability in Prisons Project. ................................................................................................. 39
Figure 14. Linear regression output of the average temperature and average relative humidity
versus life stage length for the egg life stage. ............................................................................... 44
vi

Figure 15. Boxplot and histogram for GDD egg life stage. In 2021, accumulated GDD min =
151, median = 255, max = 340. In 2022, accumulated GDD min = 147, median = 239, max =
314................................................................................................................................................. 44
Figure 16. Linear regression output of the average temperature and average relative humidity
versus life stage length for the 1st instar stage.............................................................................. 46
Figure 17., Boxplot and histogram for GDD 1st instar life stage. In 2021, accumulated GDD min
= 96, median = 175, max = 268. In 2022, accumulated GDD min = 66, median = 198, max =
379................................................................................................................................................. 46
Figure 18. Linear regression output of the average temperature and average relative humidity
versus life stage length for the 2nd instar life stage....................................................................... 47
Figure 19., Boxplot and histogram for GDD 2nd instar life stage. In 2021, accumulated GDD
min = 11, median = 22, max = 218. In 2022, accumulated GDD min = 11, median = 128, max =
350................................................................................................................................................. 48
Figure 20. Linear regression output of the average temperature and average relative humidity
versus life stage length for 1st instar to 3rd instar duration. ......................................................... 49
Figure 21. Boxplot and histogram for GDD hatch to 3rd instar life stage. In 2021, accumulated
GDD min = 190, median = 274, max =367. In 2022, accumulated GDD min =197, median 324,
max = 506. .................................................................................................................................... 50
Figure 22. Linear regression output of the average temperature and average relative humidity
versus life stage length for the egg to 3rd instar duration. ............................................................ 51
Figure 23. Linear regression output of the average temperature and average relative humidity
versus life stage length for the 3rd instar to diapause duration. ................................................... 53
Figure 24. Boxplot and histogram for GDD third to 5th instar life stage. In 2021, accumulated
GDD min = 0, median = 315, max = 428. In 2022, accumulated GDD min = 261, median = 389,
max = 701. .................................................................................................................................... 53
Figure 25. Linear regression output of the average temperature and average relative humidity
versus life stage length for the egg to diapause duration. ............................................................ 54

vii

LIST OF TABLES

Table 1. Parameter estimates (effect size) for average temperature and average relative humidity
by development period per year. All reported effect sizes with a standard error (SE) are
statistically significant (p < 0.05). NS = no significant relationship. ......................................... 55
Table 2. Hatch rate, egg estimates, and larvae counted at 3rd instar by period per year. .......... 55
Table 3. Accumulated GDD values by life stage period and quartiles per year. ......................... 55

viii

ACKNOWLEDGEMENTS

I could not have made it this far without the tireless help of John Withey, my reader and
professor, who dedicated an impressive amount of time and patience in assisting me weekly with
everything from statistical analysis to thorough feedback on this research. I am also in
appreciation to MES director Kevin Francis and assistant director Averi Azar, for supporting me
throughout my journey as a graduate student.
This research would not have been possible without the support of the other partners and
individuals connected to this butterfly. I extend my deepest gratitude to the 2022 and 2023
members of the Sustainability in Prisons Project, including co-director Kelli Bush, whose
support and guidance over this past year and a half has been immeasurable. I thank the partners
of the Sustainability in Prisons Project, including The Evergreen State College, Department of
Corrections, and Washington State Fish and Wildlife, whose engagement and collaboration
continue to make the butterfly program possible. I also thank the individuals within these
agencies, including: biologists Mary Linders and Andy Dechaine from WDFW; butterfly
program manager at Coffee Creek Corrections Center Ronda Naseth; CPM Courtney Robbins,
Ms. April Henderson, and butterfly program liaison Ms. Dalynn Martinez from DOC.
The butterfly program at Mission Creek Correctional Center for Women also has a unique
tradition of the passing down of knowledge among previous butterfly coordinators and butterfly
technicians. Thus, I express my gratitude for the previous butterfly coordinators before me and
past butterfly technicians, who laid the groundwork for this research. I also extend my
appreciation to Dennis A. Buckingham, who first introduced me to the Taylor’s checkerspot and

ix

encouraged me to connect with SPP, and Fayth Shuey, who’s leadership played a role in my
forming a deep connection to the butterfly’s native prairie habitat.
I also extend my respect and deepest gratitude to Dr. Michael R., who has provided me with
wisdom and counsel insurmountable to my growth and progress throughout this experience.
Lastly, I owe my deepest appreciation to the team of incarcerated butterfly technicians during the
2022 and 2023 years; for their data collection that this thesis is built upon, and their overall
dedication to the scientific work they do to save this butterfly. This research would not be
possible without you.

x

ACRONYMS
TCB – Taylor’s checkerspot butterfly
JBLM – Joint base Lewis-McChord
MCCCW – Mission Creek Corrections Center for Women
CCC – Coffee Creek Corrections Center
OR – Oregon
SPP – Sustainability in Prisons Project
WDFW – Washington Department of Fish and Wildlife
DOC – Department of Corrections

xi

Introduction
Prairie and oak-woodland ecosystems were once one of the more dominant ecoregion
types in the Pacific Northwest lowlands, maintained through Native American burning and
wildfires (Bachelet et al., 2011). Euro-American settlement occurred around the 1840s in the
Pacific Northwest, and the process of colonization resulted in dramatic alterations of the prairiesavannah landscape. The genocide and cultural assimilation of Native Americans by EuroAmericans in the Pacific Northwest and governmental emphasis on fire suppression disrupted the
natural prairie-savannah burn cycle (Bachelet et al., 2011; Martin & Kronland, 2015).
Additionally, Euro-American agricultural practices created habitat loss and fragmentation which
drastically shrank the prairie landscape. Prairies are now classified as one of the rarest
ecosystems of North America and have been reduced to only 3% of what existed prior to EuroAmerican settlement (Bachelet et al., 2011; Martin & Kronland, 2015).
With Pacific Northwest prairie-savannahs imperiled, these ecoregions may be especially
sensitive to the negative effects of climate change. Current trends predict that the Pacific
Northwest will overall become warmer, with increased risk of droughts during the summers and
more variable precipitation (USGCRP, 2018). Though prairie-savannas are a fire-dependent
ecosystem, an increased risk of wildfires and hotter summers may cause more harm than provide
benefits. Climate change is projected to lengthen growing seasons, fire seasons, reduce moisture
in the soil, and increase fire severity and area burned (McKenzie et al. 2004; Westerling, 2016;
Halofsky et al., 2020). Fire that is too frequent, intense, or severe runs the risk of sterilizing soils,
causing native perennials to struggle, and negatively impacting pollinator presence (Adedoja et
al., 2019; Potts et al., 2003). The changes of fire regimes within Pacific Northwest prairies in

1

addition to shifting environmental conditions and growing season length can in turn drastically
affect the several endangered species only found within Pacific Northwest prairie ecosystems.
One of the most sensitive groups of organisms to temperature and precipitation effects of
climate change are butterflies, which function as indicators of both climate change and overall
environmental health (Vickery, 2008). Butterflies have a co-evolutionary relationship with plants
and are vital to plant pollination (Janzen, 1980; Vickery, 2008). A greater abundance of
butterflies found within an ecosystem is indicative of good ecosystem health, and increased
populations of butterflies may indicate increased diversity in plant communities, in addition to
the possible presence of other pollinators (Ghazanfar et al., 2016). Butterflies also are a food
supply for other organisms and may pass plant chemical properties and toxins onto higher
trophic levels. The Taylor’s checkerspot butterfly is of special concern to Pacific Northwest
researchers and conservationists; this species has been listed as federally endangered since 2013,
and declines in their populations foreshadow the possible loss of the entire glacial outwash
prairies they inhabit (Bachelet et al., 2011; Hill & Martin, 2019; Murphy & Weiss, 1988).
Research on the influence of environmental conditions on checkerspots is scarce. Ehrlich
et al. (1980) studied the effects of the 1977 California drought on Californian Euphydryas
populations and found that drought stress on plants lead to harsh declines in these populations.
Parmesan et al. (2015) found that climate change was driving range shifts for the Quino
Checkerspot, particularly temperature. Hill et al. (2017) examined the relationship between
weather and prescribed fire fuel conditions and their impact on butterfly communities in the
Puget Sound area. Bennett et al. (2014) analyzed the effects of microclimate on Taylor’s
checkerspot habitat use, larval distribution, and adult butterfly movement in situ within Oregon

2

State. However, the association of environmental conditions and growing degree days to
checkerspot development and life stage length has gone generally unexplored.
Due to conservation concerns, captive rearing programs for the butterfly have surfaced
over the last two decades. The Oregon Zoo began a captive rearing program in 2003, in which
zoo staff developed the first husbandry manual for the butterfly and raised it through its life
stages (Leroy, 2012). This in part helped to initiate the establishment of the butterfly program at
Mission Creek Correctional Center for Women (MCCCW) in Belfair, Washington, founded by
the Sustainability in Prisons Project in collaboration with the Department of Corrections (DOC)
and Washington State Department of Fish and Wildlife (WDFW) in the 2011 and 2012 years
(Leroy, 2012). A third butterfly program was then initiated at Coffee Creek Correctional Facility
(CCCF) in Corvalis, Oregon, in 2017. The establishment of these programs provide an
opportunity to study the relationship of temperature, relative humidity, and GDD of different
instar stages in a captive rearing setting, providing data and observations that can then be
compared with wild populations of the butterfly and controlling for genetics, since butterfly
matrilines and their eggs and larvae could be tracked and considered separately.
This thesis research seeks to provide an understanding of temperature and humidity on
the length of individual instar stages for the Taylor’s checkerspot butterfly using data collected
between 2021 and 2022 years from the Sustainability in Prisons Project’s (SPP) captive-rearing
program at Mission Creek Correctional Facility for Women (MCCCW). Growing degree days
(GDD) was also analyzed to help understand whether it could be used successfully as a metric to
potentially predict butterfly phenology for the Taylor’s checkerspot in future studies. This
research may be used to help better inform program decisions made at the butterfly program at

3

MCCCW and contribute to scientific knowledge surrounding climate change and the effects on
pollinator species.

4

Literature Review
The Taylor’s checkerspot butterfly (TCB) is endemic to Pacific Northwest prairies and
balds (dry lowland sites dominated by herbaceous plants and dwarf shrubs) within Vancouver
Island, Oregon’s Willamette Valley, and the Olympic Peninsula (Potter, 2016). Once described
to “swarm by the thousands”, Euro-American urban development, disruptions to Native
American burning, and the introduction of exotic grasses have since resulted in local extinctions
of Taylor’s checkerspot populations (Severns & Warren, 2008). Several agencies and
organizations have undertaken efforts to increase Taylor’s checkerspot survival in Washington
State, including organizations that focus on captively rearing the butterfly. To better help inform
program decisions surrounding captive-rearing, growing degree days (GDD) may be used. GDD
has the potential to help predict when the different Taylor’s checkerspot life stages will occur
and provide information on the effects that accumulated heat have on Taylor’s checkerspot
development.
For this literature review, since butterflies are typically referred to by their common
names, “checkerspot” may be used in place of Euphydryas, and the Taylor’s checkerspot
butterfly will be shortened to its commonly used acronym “TCB”. Additionally, there are several
common definitions of the “Pacific Northwest,” none of which are universally accepted
(Richards, 1981). For this review, I define “Pacific Northwest” as the geographic region that
encompasses British Columbia, Washington State, and Oregon. This review will begin with a
description of GDD and its general uses before moving on to provide an overview of TCB.
Effects of environmental conditions on checkerspots, phenological asynchrony of host plants,
and current conservation measures will be the primary focus of this literature review.

5

Growing Degree Days (GDD) – General uses
Growing degree days (GDD) is a unit of measurement describing total annual heat
accumulation above a specific temperature threshold (Cayton et al., 2015). GDD has been used
in agricultural research since the 1980s to predict the timing of key events within plant and insect
lifecycles (Parry & Carter, 1985; as cited in Cayton et al., 2015). Plant and insects have a total
heat requirement for each developmental stage, and GDD is used to predict an organism’s length
of life stage by calculating accumulated degree days (Bonhomme, 2000; Miller et al., 2001).
Predicting an organism’s development can be useful to know in order to treat for pests and
allows for better crop management (Miller et al., 2001). Additionally, exceeding a certain
accumulated heat threshold may slow or prevent an organism’s development.
Over the years the concept of using GDD in phenological studies has extended from the
field of agriculture to ecology, and has been used throughout research as a predictor for butterfly
species development, abundance, and distribution (Hellegers et al., 2022). Cayton et al. (2015)
found that GDD was a better tool for predicting butterfly phenological changes as opposed to
calendar dates. Bristow et al. (2022) found that number of degree-days needed for triggering the
hatching and eclosion of the endangered Karner blue butterfly was decreased under warming
temperatures, indicating that development was accelerated for these life stages. In other studies,
accumulated degree days were used to predict early flight periods for the Michigan butterfly and
skipper populations (Perkins 2007), and growing degree days had greater ability to predict
butterfly dispersal in boreal ecosystems than other variables (Luoto et al., 2006).
Although GDD has been used in butterfly research and the Taylor’s checkerspot butterfly
is a federally endangered species, GDD has not been studied within Taylor’s checkerspot
butterfly populations. Understanding how total heat accumulation relates to checkerspot life
6

stage length and development in captivity can provide several benefits to agencies, researchers,
and other organizations supporting checkerspot conservation efforts and prairie ecosystems. The
potential for predicting the timing of key life stages may be used to help guide husbandry
methods and program management decisions, such as preparing for larval releases or
accommodating care for various instar stages. GDD may also help to inform how the butterfly is
enduring climatic changes, and perhaps be compared to GDD research on its primary host plant,
Plantago lanceolata, to identify phenological asynchrony.

Taylor’s Checkerspot Butterfly – General Overview
Euphydryas editha
The Taylor’s checkerspot butterfly is a subspecies to Euphydryas editha. Euphydryas
editha is a univoltine (producing one generation in a season), non-migratory species of butterfly
occurring in populations or metapopulations throughout the Pacific Northwest United States and
Canada (Bennett et al., 2015). The flight season of E. editha lasts anywhere from three to five
weeks ranging from March through July, and eggs generally hatch after two weeks of being laid
(Bennett et al., 2015). New larvae generally stay near where they hatch and generally begin
feeding on the plant they hatch from, therefore, mother butterflies finding a suitable host plant in
ideal habitat is important (Bennett et al., 2015). The plant families Plantaginaceae and
Orobanchaceae offer suitable candidates as hosts, with Plantago, Castilleja, and Collinsia being
the most popular choices (Bennett et al., 2015; Buckingham et al., 2016; Dunwiddie et al., 2016).
Larval development and behavior
A significant portion of checkerspot research was undertaken by Paul Ehrlich, who
studied Euphydryas editha taylorii extensively throughout his career. He published On the Wings
of Checkerspots: A Model System for Population Biology with researcher Ilkka Hanski in 2004.
7

Their work has greatly informed scientific understanding of E. editha taylorii, including larval
development and behavior. As described within this book, Taylor checkerspots lay their eggs in
clusters of anywhere between 5-500 eggs during late springtime, though at the butterfly program
at Mission Creek Correctional Facility some butterflies have been observed to lay as many as
800 eggs (personal observation). This is a unique characteristic to the species since 90-95% of
members in the Lepidoptera order lay eggs singularly. Laying their eggs in clusters may put
checkerspots at greater risk for extinction since egg clusters are more prone to population
fluctuations (Ehrlich & Hanski, 2004).
When checkerspots hatch, the larvae live in groups and molt through five instar stages
prior to entering diapause together, which is an immobile period of suspended development that
occurs extreme temperatures, usually during the fifth instar stage (Ehrlich & Hanski, 2004).
Taylor’s checkerspot larvae consume their host plants as a food source, which is mostly either
Plantago lanceolata or Castilleja spp. After diapause larvae molt into a sixth instar stage, and
then eventually pupate (Ehrlich & Hanski, 2004). Checkerspot caterpillars are only able to move
short distances from their original host plants and spin webs, which may provide them with
further protection during diapause (Buckingham et al., 2016; Ehrlich & Hanski, 2004; Stinson,
2005). A proportion of larvae within a population may re-enter diapause multiple times and
remain in the larval life stage for years if conditions are unfavorable (Ehrlich & Hanski, 2004).
Beginning in the third instar stage, larvae are black and have brightly orange-colored dots
along their backs. Larvae consume and sequester iridoid glycosides (plant chemical compounds)
from their host plants, which provide them with a chemical defense against bird prey and
arthropods such as spiders, making then unpalatable to these predators both in their larvae and
adult life stages (Ehrlich & Hanski, 2004). However, the extent of this unpalatability has been
8

observed to be variable in populations. Though checkerspots have developmental benefits that
aid to their survival—such as multi-diapausing and sequestering iridoid glycosides—other
environmental factors have resulted in their endangerment.

History and Endangerment Status
The Puget lowland prairie landscape was formed by a 3,000-foot glacier known as the
Puget lobe of the Cordilleran Ice Sheet entering the area roughly 17,600 years ago, melting
approximately 2,700 years later (Williams, 2016). This glacier deposited softer sediments as it
made its way through the lowland area, creating areas with practically no hard rock. Currents of
water beneath the Puget lobe as it melted, in addition to the ice movement, sculpted this sediment
layer, and both these glacial sediments and the currents formed the topography of the land
(Williams, 2016). Glacial deposits of these sediment types within Taylor’s checkerspot habitat
have resulted in loamy and well-draining prairie soils with high organic content. In addition to
this glacier and its currents carving the landscape and depositing sediments, after the Puget lobe
melted, the unique vegetation of prairie ecosystems in the Puget Sound was influenced by
Indigenous “fire economies” within the region, until the introduction of Euro-American fire
suppression (Boyd & Lake, 2021).
European settlers and Native tribes in the Pacific Northwest made contact as early as
1774, with most early European colonization activities beginning in 1812 when fur trading was
established (University of Washington, n.d.). Early European explorers described the prairies of
the Willamette Valley in Oregon and areas in Victoria that Indigenous people shaped with fire as
being “open”, and because of this openness was selected as the central site for the Hudson Bay
Colony in 1845 (Boyd & Lake, 2021). Indigenous use of fire in the Pacific Northwest was
primarily used to help source food commonly found on the same prairies home to the Taylor’s
9

checkerspot butterfly, such as camas and wild berries. Within the Nisqually region specifically,
which is within the area of Joint Base Lewis-McChord—where the butterfly survives today—
fires were noted to have been very patterned. Burns seemed to fall between August 13 and
September 12, with some sporadic fires prior to this date (Boyd & Lake, 2021). Suppression of
Indigenous burning by Euro-Americans began around the early 1900s, and eventually led to the
suppression of fire entirely (Bachelet et al., 2011; Boyd & Lake, 2021). Fires from Indigenous
burning were critical to Taylor’s checkerspot habitat, preventing the encroachment of conifers
and nourishing native nectar and host plants used by the butterfly (Hill et al., 2017; Hill &
Martin, 2019).
Due to urban expansion being more convenient on prairies than mountainous landscapes,
the Puget lowlands has become one of the most densely populated areas of Washington State.
Prairie grassland vegetation has been reduced by 90% of its original landscape, and assessments
conducted from 1994-1995 showed significant degradation and loss of prairie habitat on Joint
Base Lewis-McChord due to both military use and urban development (Crawford & Hall, 1997).
Due to fire suppression and urban expansion, checkerspot habitat fragmented and their
populations decreased over time. If species are unable to move within fragmented habitats, their
populations become isolated and are less resilient to disasters, unfavorable conditions, or drastic
changes to their environments (Bennett et al., 2013). It is common for butterflies to exist as
metapopulations (collections of subpopulations), and the Taylor’s checkerspot butterfly
specifically does not disperse long distances, making it especially difficult to recover from local
extinction events (Stinson, 2005).
Another influencer to Taylor’s checkerspot habitat fragmentation was the advent of
European agricultural practices and invasion of exotic grasses, which drastically degraded
10

Pacific Northwest prairies, further reducing the resilience of these metapopulations (Bachelet et
al., 2011; Bennett et al., 2013; Buckingham et al., 2016). Exotic grasses quickly take over the
native prairies of Oregon’s Willamette Valley and Joint Base Lewis-McChord, which are also
occupied by the butterfly (Poulos & Roy, 2015; Severns & Warren, 2008). The TCB primary
host plant Plantago lanceolata is quickly crowded out by other taller invasive grasses, and
flowers of Fragaria virginiana, a primary nectar source for E. taylori, also become outcompeted
(Severns, 2008). Anthropogenic climate change provides invasive species that are already
outcompeting checkerspot resources with further advantages, which could be devastating to
checkerspot metapopulations. Hotter summers caused by climate change will increase the
frequency and intensity of wildfires, and in turn these fires perpetuate the spread of taller
invasives, which usually grow more quickly after burns (Poulos & Roy, 2015). This results in a
grass-fire cycle, wherein invasive grasses create favorable, low-intensity fire conditions that
further benefit their seeding and quick establishment (D’Antonio & Vitousek, 1992).

Phenological asynchrony and mismatch
The timing of key events during plant lifecycles (phenology) has been disrupted due to
climatic changes (Reed et al., 2019). These changes can result in negative outcomes for
pollinator species. Temperature is a major control of phenology, triggering the budding of new
leaves and earlier production of fruits and seeds when it is warmer (Moore, Lauenroth, Bell, &
Schlaepfer, 2015; Rathcke & Lacey, 1985; as mentioned in Reed et al., 2019). Warming
temperatures have also been found to be more impactful on plant phenology within Pacific
Northwest prairie habitats than either soil moisture or precipitation (Reed et al., 2019). An
increase in temperature has been demonstrated to delay the timing of wet seasons and either
prolong or initiate drought within prairie systems, which in turn has advanced flowering dates
11

(Reed et al., 2019). As this trend continues, phenological asynchrony may be at increased risk of
being amplified for the Taylor’s checkerspot butterfly and its host plants, since this species is
found within prairies and grasslands.
Checkerspot butterflies already have a natural disadvantage in the face of climate change
because of their natural asynchrony with its host plants. The Bay checkerspot butterfly, another
subspecies within E. editha and native to California, has been shown to experience severe losses
of its larvae due to starvation, with 90% larval mortality during the late springtime (Weiss et al.,
1988; Singer, 1972; as mentioned in Parmesan et al., 2015; Singer & Parmesan, 2010). These
heavy losses are a result of the early senescence (dying) of its host plant Plantago erecta prior to
the larvae being healthy enough to enter diapause for the hotter summer season (Harrison et al.,
1988).
This asynchrony between E. bayensis and Plantago seem to have existed before changing
climatic conditions, indicating that larval mortality may be an adaptive tradeoff for adult
fecundity (maximum reproductive potential) (Weiss et al., 1988; Singer, 1972; as mentioned in
Parmesan et al., 2015; Singer et al, 2010). As seasonal warming trends increase and key events
within Plantago’s life cycle happen earlier and earlier, the asynchrony that already exists
between E. bayensis and Plantago is likely to be exacerbated far beyond the ability for the Bay
checkerspot to adapt. In other words, the Bay checkerspot butterfly is already considered to be at
its ecological threshold for its ability to tolerate this asynchrony, and further increases in
asynchrony may result in a phenological mismatch (Renner & Zohner, 2018; Singer &
Parmesan, 2010) with negative consequences for the Bay checkerspot.
In E. editha species including the Taylor’s checkerspot, pre-diapause larvae that come
from females who fly the earliest tend to have higher survivorship than those descended from
12

later flying females (Weiss et al., 1988). This is because female butterflies that fly the earliest lay
their eggs on north-facing slopes, which tend to be cooler and grow plants that senesce later, so
these larvae can feed on these plants for longer (Weiss et al., 1988). However, the early-flying
female butterflies that these pre-diapause larvae descend from develop as larvae on south-facing
slopes, which have higher temperatures and higher starvation rates for pre-diapause larvae, and
therefore fewer survivors (Weiss et al., 1988). Under climate change, this dichotomy is likely to
be further disrupted by increasing temperatures. This is because temperatures on both the cooler
north-facing and south-facing slopes will rise from global warming, so host plants on the cooler
north-facing slopes would senesce earlier, thereby resulting in less food availability for
checkerspot larvae.

Phenological changes in butterflies under climate change
In addition to asynchrony with host plants, the effects of climatic changes have resulted
in changes within butterfly development. As ectotherms (a term that describes organisms which
obtain heat from their environments), butterfly development is heavily dependent upon
temperature (Forrest, 2016). The rate at which a butterfly develops through a given life stage—
including checkerspot butterflies—increases when temperatures are warmer. Warmer spring
times have been observed to result in earlier eclosion (emergence from chrysalis) and flight
periods for butterflies (Forrest, 2016; Roy et al., 2015).
Male Taylor’s checkerspot butterflies also tend to eclose earlier than females, and
research seems to be limited on whether increased temperatures may affect sex bias (greater ratio
of one sex in comparison to the other) of the organism in the field during the flight period.
Additionally, some butterfly species have also shown a shift in previously being a univoltine

13

(producing a single brood of offspring per year) to becoming multivoltine under warming
temperatures, with multivoltine species producing more generations per year (Altermatt, 2010).
In contrast, increased precipitation has been demonstrated to lengthen the life stage of
pupation and delay eclosion, but too much rainfall can also create a delay for post-diapause
larval development (Lagerquist, 2019). This may result in a shorter flight period for adult
females and a smaller window for them to oviposit prior to host plant senescence (Lagerquist,
2019). Additionally, cooler temperatures and greater humidity during the egg period seem to
indicate a longer duration of time spent in the egg stage (Severns & Grosboll, 2011).
Precipitation patterns have been predicted to change within the Pacific Northwest region based
on global projections, and under climate change increased temperatures and greater or more
sporadic rainfall may disrupt flight periods for Taylor’s checkerspot butterflies, result in shorter
developmental windows for life stages, and potentially result in heavy sex bias during flight
season (USGCRP, 2018).

Current Conservation Measures
Institute for Applied Ecology and Oregon Zoo efforts
The first captively-reared Taylor’s checkerspot butterfly within the Pacific Northwest
started accidentally in 2003, after a Washington Department Fish and Wildlife voucher specimen
originating from Clallam County began to oviposit on a species of Castilleja (Schultz et al.,
2011). Following their establishment of Oregon silverspot butterfly (Argynnis zerene hippolyta)
captive rearing in 1999, the Oregon Zoo initiated a captive rearing program for the Taylor’s
checkerspot butterfly in 2004, collecting wild eggs from field locations and raising them to prediapause larvae wherein they were released back out into the field. The collection of eggs was
limited to 1-2% of the butterfly’s estimated population, which translated to no more than 600
14

eggs (Shultz et al., 2011). Much of the groundwork science in Taylor’s checkerspot captive
rearing began under the care of Oregon zoo staff, with guidance for the development of protocols
created in collaboration with Dr. Gordon Pratt (Shultz et al., 2011).
Beginning 2008 the Oregon Zoo successfully began to rear larvae through to the adult
stage and initiated captive breeding. Wild adult females were also collected and brought into the
zoo in 2008 to allow for morphometric data to be gathered and compared between captively bred
and wild-caught populations (Shultz et al., 2011). This proved to be an ideal time, since the
following year the population of Taylor’s checkerspot butterflies at the field collection site
sharply declined, and out of concern for the strain placed on the population biologists refrained
from collecting wild eggs and adults. This same year, the zoo published a captive rearing
husbandry manual in 2009 for the butterfly, further helping to inform captive rearing practices
and techniques (Barclay et al., 2009; Shultz et al., 2011).
Despite the 2009 plunge in population numbers, captively raised butterflies manage to
produce over 10,000 estimated eggs and 8500 larvae the following year, far exceeding the zoo’s
capacity, with these larvae being released at several different site locations at the pre-diapause
stage. Ideal spring weather conditions from spring 2009 to the following spring in 2010 allowed
for a greater flight period and successful copulations among adults released at these sites (Shultz
et al., 2011). In the year of 2020, the Oregon zoo stopped its captive rearing program at the
onset of the COVID-19 pandemic (Ronda Naseth, personal communication).
In 2016 the Institute for Applied Ecology created a 9 year action plan for the Taylor’s
checkerspot butterfly in Oregon. The nonprofit’s goal was to restore habitat and improve habitat
quality for the butterfly, increase butterfly population numbers, and assist in providing
information for other checkerspot recovery plans (Menke & Kaye, 2016). Restoring and
15

maintaining suitable habitat, increasing TCB numbers and populations and creating a
metapopulation structure, and contribute to TCB recovery planning were goals established and
still presently undertaken to support checkerspot conservation work under this plan (Menke &
Kaye, 2016).
Sustainability in Prisons Project and MCCCW
The Sustainability in Prisons Project (SPP) was formed in 2004 as a partnership between
the Department of Corrections and The Evergreen State College with the goal of bringing
science education into prisons and reducing the economic, environmental, and human cost of
prisons through sustainable initiatives. Initially beginning as a lecture series, SPP inspired
several sustainability projects at Cedar Creek Correctional Center, and the success of these
projects lead to other sustainability programs establishing within other Washington State prisons
(Aubrey, 2013).
From 2011-2012, ten incarcerated women were hired to be the first butterfly technicians
for the first year of SPP’s butterfly program. The program initially relied heavily upon the
butterfly husbandry manual developed by the Oregon Zoo, and training in addition to weekly
support was provided by an Evergreen State College graduate student enrolled in the Master of
Environmental Studies (MES) program, and employed as SPP’s butterfly program coordinator.
Biologists from the Oregon Zoo also occasionally visited the butterfly program to provide
guidance. The painted lady butterfly (Vanessa cardui) was used as the original training candidate
for butterfly technicians and the butterfly coordinator to practice husbandry rearing techniques
prior to working with the federally endangered Taylor’s checkerspot butterfly.
In March 2012, 755 Taylor’s checkerspot larvae were provided to the SPP butterfly
program by staff at the Oregon Zoo. These larvae were raised successfully, with 600 of them
16

being released onto the JBLM prairie, and the rest retained for captive breeding. Another 3395
larvae were raised as offspring from the original 155 larvae retained for breeding (Leroy et al.,
2012). During these first two captive rearing years, MES graduate student Dennis Buckingham
(previously Dennis Aubrey) conducted a thesis research project examining Taylor’s checkerspot
host plant preference, which would go on to later be published (Aubrey, 2013; Leroy et al.,
2012).
Today, SPP is either directly or indirectly involved in the establishment of sustainable
programs within every Washington State prison, including captive rearing and other conservation
efforts for the Taylor’s checkerspot. Native and rare plant propagation operations were started at
Stafford Creek Corrections Center (SCCC) and Washington Corrections Center for Women
(WCCW), which produce native plant species to be planted in Taylor’s checkerspot sites on
JBLM to help restore butterfly habitat. In 2012 a new SPP program guided by the Washington
State Department of Fish and Wildlife began at Mission Creek Correctional Center for Women
(MCCCW) in Belfair, wherein incarcerated women directly captively raised Taylor’s
checkerspots to be released at these locations.
The SPP butterfly program has also been a catalyst in supporting Master of
Environmental Studies (MES) graduate student thesis research on the butterfly. Dennis
Buckingham was the first butterfly program coordinator for SPP, and the first graduate student to
conduct research on the butterfly in SPP’s captive rearing program in 2013 (Leroy, 2012). His
research focused on Taylor’s checkerspot oviposition host plant preference and found that the
butterfly preferred Castilleja levisecta and Castilleja hispida over Plantago lanceolata (Aubrey,
2013). Wendy Lagerquist in 2019 focused her research on winter temperature, humidity, and
precipitation impacts on peak abundance of the butterfly across three different sites, finding
17

weak associations with winter climate variables on some of these sites while discovering that
peak abundance increased across these years, potentially due to checkerspot reintroduction
efforts (Lagerquist, 2019). Keegan Curry analyzed ten years’ worth of reproductive data for the
butterfly at MCCCW and CCCF captive rearing facilities, finding that the Taylor’s checkerspot
produces less offspring than other Euphydryas editha subspecies (Curry, 2019). Carly Boyd
examined how often environmental targets at MCCCW were achieved over seven captiverearing years, and whether falling outside of these environmental targets impacted larval
survival, hatch rate, fecundity, and other metrics. Boyd found that meeting these conditions was
highly variable over the years, and that achieving environmental targets did not necessarily
impact developmental milestones and captive rearing program goals for the butterfly (Boyd,
2021). Ultimately, MES students have studied different metrics and influences on both captively
reared and wild butterflies, with several of these individuals utilizing SPP butterfly program
captive rearing research.
Coffee Creek Correctional Facility and Oregon Fish & Wildlife
A Taylor’s checkerspot captive-rearing program was launched in 2017 at Coffee Creek
Correctional Center Facility (CCCF), a women’s prison in Oregon State. This butterfly program
was the first butterfly conservation program to be introduced into a medium security facility, and
modelled the SPP butterfly program at MCCCW, receiving funding from U.S. Fish and Wildlife
Service (USFWS) and the Oregon Zoo (Oregon Zoo, 2019). Chad Naugle from Oregon
Department of Corrections advocated for the program to be implemented due to wanting a
meaningful work and educational opportunity to be available to adults serving longer sentences,
and initially establishing the program was challenging due to the vast majority of conservation
programs existing within minimum security facilities (Ronda Naseth, personal communication).
18

Ronda Naseth, a zookeeper at the Oregon Zoo, became employed as CCCF’s program
manager, and her two years of butterfly husbandry training for the zoo in addition to the Oregon
Zoo’s butterfly husbandry manual was relied upon at the program’s inception. Naseth modified
the Oregon Zoo’s procedures to allow butterfly technicians at the Coffee Creek program to work
more independently, collaborating with Oregon Zoo staff to supplement visits to the facility and
receive answers to any inquiries (Ronda Naseth, personal communication). While for the most
part butterfly technicians work independently, Ronda plays a central and important role in
supporting the CCCF butterfly team and providing them with guidance.
Unlike the SPP butterfly program area, which consists of two greenhouses and a shed
outside of the prison perimeter fence, CCCF’s butterfly lab exists within the medium security
facility and is connected to one of the secure housing units. The room that is now the butterfly
lab previously was used as a classroom and for overflow housing, and additions had to be made
to the infrastructure for the lab’s creation, including plumbing, filtration, a sink, and an HVAC
unit. Longer counters and shelving units were also built within the room to allow for better
workstations and storage. Two windows overlook the yard, and the CCCF’s diapause shed rests
outside underneath one of these windows (personal communication, Ronda Naseth). While there
were several modifications needed to be made both in the infrastructure and how work processes
were done in order to meet security standards, ultimately the program has been a success in
allowing for both incarcerated women to receive essential work skills within the sciences and a
butterfly conservation lab to be tailored to captively rearing Oregon Taylor’s checkerspots.
CCCF’s butterfly lab has also invested important knowledge and scientific observations
pertaining to mortality causes of the Taylor’s checkerspot during the wake-up stage, since this
butterfly lab has experienced increased mortality numbers in its animals during this time. The lab
19

had its plantain quality assessed and ruled out as a cause for these mortality trends, and the
butterfly technicians under the guidance of Ronda Naseth conducted a light study and introduced
a limited amount of UV lighting into the lab. While this study is still ongoing, the team
hypothesizes that UV light plays a significant factor in larval survivorship.
The butterfly program at CCCF also had a surplus of larvae during the 2021-2022 captive
rearing season, which resulted in the creation of a backup lab for 2000 of these larvae and the
CCCF butterfly technicians and butterfly program manager observing a potential correlation with
length of time spent as pre-diapause larvae and survival at wake up when temperatures in the
back up lab decreased lower than CCCF’s original butterfly lab. Naseth observed larvae in the
back up lab spending an average of 2 days longer in the 4th instar life stage duration, and these
larvae had higher survival than CCCF larvae post-diapause. During the 2022-2023 season, CCCF
attempted to replicate these conditions at 4th instar using air conditioning, improving larvae
survivability by 10% more than their previous best season (Ronda Naseth, personal
communication).
As of the 2023 year, the Coffee Creek Correctional Center’s butterfly program has
successfully captively reared and released a total of 9,089 animals since its conception, and are
retaining multi-diapause larvae for the first time this year, with the intention of these animals
being released to a newly established Hood River Taylor’s checkerspot conservation facility
when it opens. While CCCC’s butterfly program does not breed Taylor’s checkerspot butterflies,
the butterfly technicians observed the eclosion of a butterfly for the first time in 2023 when one
was accidentally left behind during a pupae release and returned to the prison. Since the
program’s beginning, roughly 17 butterfly technicians have had experience working in the lab,

20

with 8 of these individuals having had been a part of the butterfly program for the past three
years.

Conclusion
Under climate change, landscape changes due to shifts in fire regimes and phenological
asynchrony introduce new challenges to endangered species conservation. Growing degree days
has been widely used in agriculture and implemented in butterfly research, demonstrating
promise to be a useful tool for predicting checkerspot phenology. Some research exists which
analyzes the effects of environmental conditions on Taylor’s checkerspot, but there is no current
research that analyzes temperature, relative humidity, and growing degree days as statistical
variables with potential influences on checkerspot life stage length. This study uses datasets
collected for captively-reared Taylor’s checkerspot butterflies from the Sustainability in Prisons
Project to ask how environmental conditions affect butterfly life stage length and development,
with a focus on growing degree days (GDD).

21

Methods
All research was conducted at Mission Creek Correctional Center for Women
(MCCCW), a minimum-security women’s prison located in Belfair, Washington within the
Tehuya State Forest. Captive-rearing procedures for MCCCW were developed by the
Sustainability in Prisons Project (SPP), a partnership between the Department of Corrections and
The Evergreen State College, and these procedures sometimes were adapted from husbandry
practices conducted by Oregon Zoo staff. Checkerspot developmental data and dates were
collected by a team of incarcerated butterfly technicians from the 2021-2022 and 2022-2023
captive rearing years, overseen by an Evergreen State College graduate student employed as the
butterfly program coordinator by SPP.

Figure 1. Evergreen State College graduate students and SPP coordinators Raychel Dunning
and Jen Bass assisting butterfly technicians in butterfly husbandry care.

22

Site
The captive-rearing life stages for the butterfly occurred within two greenhouses at
MCCCW built and designed for the purpose of Taylor’s checkerspot captive-rearing, and these
greenhouses are located just outside the prison fence. “Raven” is the oldest greenhouse, built in
2011, and is a 24ft x 10ft glass greenhouse with UV transmitting panels. This greenhouse is
partitioned into two rooms by a glass door, with the main room 16ft x 10ft and the smaller room
8ft x 10ft. The newer greenhouse, “Turtle”, is constructed with a similar design, larger than the
Raven greenhouse. When temperatures exceed 86°F, exhaust fans, roof windows, and motorized
dampers auto-function to help regulate environmental conditions. Technicians also cover both
greenhouse roofs with 50% reflective aluminet shade cloths during mid-spring and throughout
the summer to help mitigate extreme heat. Both greenhouses also contain heating systems which
allow for heat to be adjusted both manually and automatically during colder periods.
For the diapause life stage, checkerspot larvae were kept inside an 8’ x 10’ x 8’ wooden
shed with two 2’ x 2’ windows. Both the shed and the greenhouses were built with the intention
of exposing butterflies, eggs, and larvae to ambient environmental conditions but not extreme
weather events. There are eight 4’ x 10’ grille vents for the shed, with six of these vents built into
the bottom of three walls and two near the ceiling on the walls across from the other six vents
(Curry, 2019). During the diapause stage, butterfly technicians do not regulate temperature or
humidity.

23

Figure 2. Mission Creek Correctional Center for Women’s Taylor’s checkerspot butterfly program area.

24

Figure 3. Image of greenhouses used for Taylor’s checkerspot butterfly captive-rearing and breeding. Left is Raven
greenhouse; right is Turtle greenhouse. Plantain beds near the greenhouses are covered in cloches during the
colder, winter months.

25

Figure 4. The two greenhouses and the shed are located directly next to the prison’s perimeter fence.

Environmental Data
HOBO loggers are programmed prior to deployment to capture minimum relative
humidity, maximum relative humidity, minimum temperature, maximum temperature, and
average temperature every hour for the captive-rearing year. There is an option for light intensity
data to also be captured, however this data is rarely used. The loggers are changed out generally
every 1-2 weeks during the active rearing year with 3-8 new separate loggers of the same brand
to avoid battery depletion, and the new loggers are pre-set to begin capturing data at a set time in
order to ensure that environmental data is captured continuously. During diapause, loggers are
26

changed out approximately once a month due to less need for the use of environmental data for
reporting during the winter time. The placement of the loggers attempts to emulate the current
environment of the Checkerspot’s life stage (i.e., if larvae are being kept in a plexi cup with a
folded paper towel and plantain leaves which is then placed in a shoe bin misted for humidity, so
are these loggers). The loggers are always replaced after each life stage of development is
complete. Data from the loggers are uploaded every time loggers are replaced.
The number of loggers capturing environmental data and how frequently they are rotated
out and replaced with new loggers is dependent upon the life stage. Stricter guidelines to this
process were developed and implemented in 2022, in which it was decided that at least three
HOBO loggers per greenhouse would be placed on top and mid-shelf of wire racks where
oviposition chambers are placed during the active season. HOBO loggers are kept in their own
oviposition chambers without a butterfly to attempt to capture exact conditions during
oviposition. As butterflies die and the eggs produced increase, HOBO loggers are then removed
from oviposition chambers and placed with the eggs on the egg shelves—two per rack, on one
high and one low shelf. During the third instar to diapause life stage, there continues to be two
loggers per metal rack in each greenhouse, one recording conditions on the highest shelf and one
on the bottom shelf of each rack. Depending on the number of larvae, this could mean that
anywhere from 6-8 loggers are capturing environmental data during this life stage.
After being exchanged with new loggers, the loggers previously capturing environmental
data are taken back to The Evergreen State College where their data is read out using
HOBOware software. HOBO logger data is then saved as a CSV file, and then copied and saved
as an Excel file with any data collected outside of the program area (such as during
transportation to be uploaded), deleted. The raw HOBO logger data (uploaded as a .hobo
27

filetype), CSV files, and Excel files are then uploaded and saved to SPP’s digital library where
they are then used for annual reporting purposes.

Collection of Wild Females
Female butterflies are collected by WDFW field biologists from mid-April to early May,
depending on field weather conditions. During the 2022 season, wild Taylor’s checkerspot eggs
were also collected from the field and brought to MCCCW as an emergency conservation effort
due to the concern of low butterfly numbers observed by WDFW, Oregon Department of Fish &
Wildlife (ODFW), and JBLM Fish & Wildlife butterfly surveys. These wild eggs were left on
the host plants they were found on, with the sections of the host plant containing the eggs
carefully cut away and placed inside ventilated glass specimen jars. Butterflies are typically
collected from Range 76 on JBLM, and anywhere from 40–50 butterflies are taken from the
field. During the 2022 year, butterflies were collected from Range 53 and wild eggs were
collected from the Scatter Creek South area. Adult butterflies are transported to MCCCW also
within ventilated glass specimen containers, then placed inside coolers. Once brought into the
greenhouses, these butterflies and any wild eggs are then processed.

28

Figure 5. Wild gravid females are collected in glass containers and
placed within coolers for transportation from the field to the prison.

Butterfly technicians divide the total number of butterflies and wild eggs between each
greenhouse and assign each specimen an ID following SPP Oviposition procedures. This ID
contains an abbreviation of the site the butterflies were collected from (such as “FL” for “Fort
Lewis), the year they were collected, and a number based upon the order that they were being
processed.

29

Procedures and life stages
Butterfly technicians follow procedures for each given life stage, which were originally
based upon protocols previously developed and husbandry techniques practiced by Oregon Zoo
staff within their checkerspot rearing program. As MCCCW’s butterfly program became more
established over the years, captive-rearing procedures were modified SPP and undergoing heavy
edits each year to help better suit captive-rearing techniques practiced by MCCCW relative to the
infrastructure. Procedures for the 2020–2022 years separately provide guidelines for the
oviposition, eggs to third instar, third instar to diapause, diapause, and wake-up life stages. Prior
to 2020, previous breeding procedures for post-diapause, pupation, eclosion, adult care, and
captive breeding life stages were also practiced. Captive breeding ended after the 2019 year and
was intended to permanently no longer be a function of the butterfly program, but this portion of
the program is being brought back in 2023 due to conservation concerns.

Oviposition Experimental Setup and Daily Care
Upon being processed into each greenhouse, adult female butterflies are placed into
oviposition chambers one butterfly at a time containing a robust plantain plant, a water sponge
saturated with filtered water, and a cotton ball soaked in a 1:3 honey solution placed into a bottle
cap. The oviposition chambers are covered with soft tule netting. The chambers are then labelled
with each female’s unique matriline ID. Butterfly technicians check for eggs daily by first
carefully removing the female butterfly using a Q-tip soaked in 1:3 honey solution and then
placing her underneath an upside down large (16 oz) clear deli cup with this Q-tip inside. As the
butterfly feeds, butterfly technicians search her oviposition plant for eggs. The technicians
remove any eggs found by carefully cutting off portions of the leaf they are laid on using

30

scissors, estimate how many eggs are in the cluster, and then place these eggs inside a lidded and
prepared small (5.5 oz) deli cup with a cone-shaped paper towel inside.
Technicians label these small egg cups with their matriline ID, the date they were
collected, and assign a new egg cup number based upon the date and order in which they were
collected. Matrilines are kept separately to help preserve genetic diversity and better inform
WDFW reporting. Eggs develop over 10-14 days and gradually change color over time from
bright yellow, to reddish orange, to dark purple, and their percent development is tracked on the
7th day based what percentage of eggs within each cup are changing color. Estimated number of
eggs each mother butterfly has laid per day, date eggs are laid, the egg cup number that each of
these egg clusters are placed in, egg percent development, hatch date, and 2nd instar date are all
recorded on a Female to Third Instar form per each matriline. Daily and grand total egg estimates
produced from each female butterfly are also tracked separately on an Egg Tally form. Female
butterflies are fed with the 1:3 honey solution and their water sponges are changed daily, and
they continue producing eggs until they die of natural causes.
Following the current oviposition procedures as produced by the Sustainability in Prisons
Project (Bush, 2022a), six HOBO loggers (three per greenhouse) are kept in a butterfly-free
oviposition chamber set up as pictured in figures 4 & 5, mimicking the environmental conditions
of the adult butterflies. For this study, the end of the oviposition life stage for each given
matriline is considered to be completed after a female butterfly has laid her last egg, and this
information can be found on both the egg tally sheets and Female to Third Instar forms
documented by the butterfly technicians. Oviposition as a life stage is considered to be
completed when all butterflies in both greenhouses have died.

31

Figure 6. Oviposition chamber with honey dome and water sponge. Image by the Sustainability
in Prisons Project.

Figure 7. Oviposition chamber enclosure with a honey dome and
soaked sponge to nourish an adult Taylor’s checkerspot butterfly.
Two HOBO loggers per greenhouse rack are placed in replicate
butterfly-free chambers. Image by the Sustainability in Prisons
Project.

32

Egg to third instar daily care and experimental set-up
The beginning of the egg to third instar life stage is considered to have begun once the
first eggs have been laid, therefore there is some ongoing overlap with the oviposition life stage.
The date in which all larvae in a small egg cup hatch form their eggs is recorded on the Female
to Third Instar form, as is the date in which all larvae within a 5.5oz cup enter into 2nd instar. 2nd
instar larvae are observed to be slightly larger than 1st instar larvae. Egg to third instar larvae are
continued to be kept inside their 5.5oz deli cups, generally undisturbed except to provide them
with a fresh Plantago leaf daily. Up to eight of these deli cups are placed inside plastic shoe bins
with blue shop towels placed in the bottom of these bins. The blue shop towels are saturated
daily to help meet humidity environmental targets.

Figure 8. Image by SPP, from SPP’s Eggs to Third Instar procedures. Eggs are cut from leaf and placed into a 5.5oz
cup, where they are then stored inside plastic shoe bins with a saturated blue paper towel.

Butterfly technicians record the dates on the Female to Third Instar form for the
following: the percent of egg development after seven days, the date that the first larvae in a cup
has hatched, the date in which the first larvae in a cup have entered into second instar, and the
date in which the first third instar molt is noticed within a cup. When all larvae within a 5.5oz
egg cup are observed to have developed into the third instar stage, the entire cup is then counted
33

and transferred into 16oz deli cups of ~15 larvae, following the Third Instar to Diapause
procedures.
If a 5.5oz cup is observed to become too overcrowded because of the presence of too
many larger third instar larvae being present within the small deli cup, but not all larvae within
this cup have molted into third instar, then an exception can be made. These third instar larvae
can be counted and transferred into larger 16oz cups following the Third Instar to Diapause
procedures, while any other larvae not yet molted into the third instar stage within the 5.5oz deli
cup remain until they, too, reach third instar and can be counted separately into bigger 16oz cups.
During this life stage, at least three HOBO loggers per each greenhouse are positioned to
mimic the conditions in which the eggs are kept following the current Eggs to Third Instar
Larvae procedures as written by SPP (Bush, 2022b). Since the HOBO loggers are larger than the
5.5oz deli cup, they are instead placed inside a 16oz cup with a paper towel and a lid. This life
stage is considered to have ended when each deli cup has been observed to have completely
developed into larger third instar larvae.

Figure 9. Photo by SPP, as found in their Eggs to Third Instar procedures .

34

Third Instar to Diapause
When all larvae within a 5.5oz egg cup are observed to have molted into 3rd instar, larvae
are transferred from these 5.5oz cups into 16oz cups one matriline at a time. Each 16oz cup
contains batches of approximately 15 larvae per matriline, with larvae from different matrilines
never being mixed together into these 16oz cups. Any 2nd instar larvae are left in 5.5oz cups until
they reach third instar and are not transferred into 16oz cups due to their fragility. The new 16oz
cups are labelled with the matriline ID, the date in which the larvae are transferred into 16oz
cups, their original hatch date, total larvae within the 16oz cup, and a unique cup ID that
incorporates the ID number on each original egg cup from a single matriline and a letter, used as
an identifier for each 16oz cup. A singularly folded paper towel is placed into 16oz cups.

Figure 10. Example of the 16oz cup label created when larvae are transferred from 5.5oz egg cups into 16oz third
instar cups. Image credit Sustainability in Prisons Project.

Butterfly technicians perform daily care for 3rd-5th instar larvae following the Third Instar
to Diapause procedures, which follow the same guidelines for daily care for the Eggs to Third
Instar procedures. Larvae are provided anywhere from 4-6 leaves daily with old leaves removed
from the cups every day. Larvae are visually assessed for good general health and mortality, and
paper towel liners in cups are changed when 25% of the towel liner has been soiled by frass or if
mold is observed. Old paper towel liners and old leaves are placed into a separate bin, then

35

double checked at the end of the day for any larvae that may have been missing. Any stray larvae
found in these bins or within the lab space are placed into a “QC” (quality control) 16oz cup
since their matriline is unknown. Like previous life stages, 16oz cups of larvae are kept inside
plastic shoe bins lined with a saturated blue shop towel, with three cups each fitting into these
bins. Bin lids are kept 3/4ths of the way closed.
When all larvae within a cup have molted into 5th instar, their cup is marked with a small
colored sticker dot with the date in which it was observed the cup had fully entered into 5th.
Technicians discern when larvae enter into 5th instar based on molts found within cups and based
upon the larvae’s slower and more inactive activity, such as frassing less, moving less, and eating
less leaves. Two weeks after the last cup of larvae within the program area has entered into 5th
instar, all larvae are allocated into diapause cups following the Diapause procedures.
Butterfly technicians record the following data on the Third Instar to Diapause form:
Matriline, number of larvae counted into 16oz cups at third instar, date counted into 16oz cups,
date in which the first larvae is observed to have molted into 4th instar, and date in which the first
larvae in each individual cup has molted into 5th instar. For this study, the 4th instar life stage is
not analyzed or used as an indicator due to the difficulty in adequately discerning the 4th instar
stage, resulting in large disparities in the data recorded. The beginning of the third instar stage is
considered to have started when larvae in a 5.5oz cup are able to be transferred into their 16oz
cups. For this research, the end of this life stage is recorded based on the date in which each
individual cup has been observed to have molted into 5th instar.
HOBO loggers for this stage are managed in the same way as for the Egg to Third Instar
life stage. Loggers are replaced at minimum once every 2-3 weeks during this life stage. HOBO
loggers are kept on the same shelves that third to fifth instar larvae are kept, within their own
36

container and in an empty shoe bin with a blue saturated shop towel along with the Min/Max
readers that butterfly technicians use to assess the environmental conditions in the greenhouse.

Diapause
Following the Sustainability in Prison’s Project’s Diapause procedures (Bush, 2021),
larvae are allocated into 16oz deli cups of approximately 50 larvae each after larvae have been
confirmed to have molted into 5th instar and show signs of diapause. Larvae are kept in their
greenhouses for two more weeks prior to being moved into the outside shed and placed
underneath terracotta pots. Three HOBO loggers are moved into their own 16oz deli cups. They
are then placed inside the shed and underneath tree separate terracotta pots, each on one rack,
without larvae underneath these pots. Technicians monitor larval movement during the diapause
period, conducting one movement check a month from September–December, every other week
for January, and then once a week for February. Larvae are considered to have “moved” if they
are observed to be outside their paper towel liners, counted, and then placed back in between
their two paper towels. Once larval movement reaches 15% or greater for two consecutive
movement checks after February 1st and WDFW has confirmed conditions out in the field to be
deemed fit for larval survival, larvae are brought back into their previously assigned greenhouses
for approximately two weeks, in which the end date for the diapause life stage and beginning of
the wake-up life stage is recorded. This stage was not included in the analysis for this thesis
research.

37

Figure 11. Image from SPP Diapause procedures demonstrating which larvae are considered “moved” during
movement checks. Image credit: Sustainability in Prisons Project.

Figure 12. HOBO loggers are placed under terracotta
pots in the outside shed during diapause to mimic the
conditions of the larvae. Image by the Sustainability in
Prisons Project.

38

Figure 13. Larvae in diapause are placed under
terracotta pots. Image credit to the Sustainability in
Prisons Project.

Data Analysis
Life stage length and development
At MCCCW, data is collected per matriline and larvae cup ID, with every matriline
having its own dedicated form per life stage. Butterfly technicians fill out these forms by hand,
and these forms are then stored in binders within the greenhouse in which the relative matriline is
kept until they are turned in to an SPP staff member at the end of the captive-rearing year. This
data is used by both SPP and WDFW to complete annual reporting and helps to inform program
decision making, measure captive rearing success, and compare with data taken from the field.
The dates of specific developmental milestones as recorded by technicians were transcribed from
these physical data forms to an Excel workbook, and then used to calculate the length of each life
stage length. The beginning and end of each life stage was analyzed per the dates written down
for each individual cup of larvae.

39

For oviposition, the life stage was considered to begin the day that each individual
butterfly laid its first eggs, and the life stage was considered to have ended the last day that each
butterfly laid eggs. The entire life stage of oviposition itself was considered “ended” after the last
butterfly within the entire program area stopped producing eggs.
For the egg to third instar life stage, this was considered to begin the given day that a
cluster of eggs per matriline was observed to have been laid by an adult female butterfly, and
then collected by the technicians and placed into its egg cup. Technicians collect eggs daily, so
“date collected” is approximately the same day the butterfly has laid eggs. Larvae then hatch
(and this date indicates the beginning of the 1st instar stage), develop into second instar, and then
third instar. This life stage was complete for an individual cup once larvae within an egg cup
reached third instar and were transferred to larger cups of approximately 15 third instar larvae.
Thus, the date that third instar larvae from a given cup were transferred to 16oz cups is used as
the end date for the egg to third instar life stage, and varied for each egg cup within each
matriline. Since the small egg cups sometimes had hundreds of eggs, if third instar larvae were
present in the cup and it was deemed to be too crowded, sometimes these larger third instar
larvae were transferred to the larger 16oz cups prior to the entire egg cup developing into 3rd
instar larvae.
The third instar to diapause life stage also begins the date larvae were transferred from
their egg cups into their 16 oz cups. Therefore, the end date of the egg to third instar life stage is
the start date of the third instar to diapause life stage. The date that the first larvae within a given
cup was observed to have molted into 5th instar marked the first day of diapause, and this was
used as the end date for this life stage.

40

The days between the start date and the end dates for each of these life stages, per each of
these cups, were then calculated using the =DAYS( ) formula in Excel.

Environmental Data
After collecting data, HOBO loggers were read out following the SPP HOBO Logger
Data Upload and Clean Up procedures provided to butterfly coordinators. Uploaded data
contained the minimum, maximum, and average temperature, relative humidity, and light data
per hour for every day. In addition to this hourly data, separate columns of the average
temperature and humidity data for each 24-hour period was also recorded.
For the 2021 year, environmental data was collected for both greenhouses from May 2,
2021 (a few days prior to the processing of adult butterflies) until May 1, 2022. For the 2022
year, data was collected from May 6th, 2022 until February 27th, 2023. However, since this
research did not encompass the cold diapause and wake up life stages, only data through
September 15th for the 2021 year and August 26th of the 2022 year was considered.
Environmental data were stored in an Excel table which included the greenhouse, year,
average temperature, and average humidity for each day in addition to logger number, life stage,
and rack placement (if this info was provided). Since there were multiple loggers per greenhouse,
the overall average temperature and humidity for each day of the rearing year across all loggers
with each greenhouse was calculated.
Using these daily averages, for a given individual cup in a specific greenhouse the
average temperature and relative humidity were calculated across the duration of the given life
stage based on its start and end date.

41

Growing Degree Days
GDD baseline of 50℉ (10℃, respectively) was used. GDD values per each given day
were first calculated within Excel, with separate columns created for each GDD baseline, by
subtracting the minimum baseline value from the given average temperature in the table for each
date and per greenhouse. Using an excel formula, GDD values were set to return an output value
of 0 if the given average temperature exceeded the baseline.
Similar to how the average temperature for each individual cup’s life stage was
calculated, growing degree days for each individual cup’s life stage was determined by taking
the sum of all GDD values for the total duration of each individual cup’s specific life stage,
based upon provided start and end dates. GDD values were not calculated for the collected to 3rd
and collected to 5th life stages due to these life stages being analyzed per matriline as opposed to
per larvae cup.

JMP
Simple linear regression was used for each given life stage to analyze relationships
between life stage duration and both average temperature and average relative humidity (across
the duration of that life stage) for individual egg and larvae cups per each year. Matrilines or cup
IDs with missing data, such as unrecorded dates, were excluded from analysis. Analyses for each
given life stage were for 2021 and 2022 years were then compared. For the collected to third and
collected to 5th life stages, average duration were calculated for each individual matriline, across
all of her egg and larvae cups. JMP was used to create boxplots for GDD, and 50F (10C) was
used as a baseline.

42

Results
Collected to hatch (egg stage)
Both 2021 and 2022 show a relationship between increased average temperatures with a
shorter duration spent in the egg stage (Figure 14). For the 2021 year, for every degree
Fahrenheit increase the duration spent in the egg stage decreased by 1.06 days, and for the 2022
year for every degree Fahrenheit increase the duration spent in the egg stage decreased by 0.59
days. There was an association with increased average relative humidity and a longer duration
spent in the egg stage for the 2021 year, in which for every percent increase in average relative
humidity the duration spent in the egg life stage increased by 0.17 days. However, no statistical
meaningful relationship was found between increased average relative humidity and egg life
stage duration for the 2022 year. The median accumulated GDD across the egg stage for 2021
was 255 and 80% of the values were between 221 and 287 GDD (Figure 15). The median
accumulated GDD across the egg stage for 2022 was 237, and 80% of the values were between
194 and 271 GDD

43

Figure 14. Linear regression output of the average temperature and average relative humidity versus life stage
length for the egg life stage. Egg cups in Raven greenhouse in red, egg cups for Turtle greenhouse in blue (n=253
individual egg cups). In 2021 and 2022, higher average temperatures for egg stage were associated with shorter life
stage duration (for 2021, F1,169 = 96.6, p <0.001, R2 = 0.36. For 2022, F1,80 = 35.3, p <0.0001, R2 = 0.31). For
2021 higher average relative humidity for egg stage was associated with longer life stage duration, for 2022 no
relationship between average relative humidity and life stage duration was found (for 2021, F1,169 = 9.82, p
<0.0020, R2 = 0.05. For 2022, F1,80 = 0.81, p =0.37, R2 = 0.01).

Figure 15. Boxplot and histogram for GDD egg life stage. In 2021, accumulated GDD min = 151, median = 255,
max = 340. In 2022, accumulated GDD min = 147, median = 239, max = 314.

Hatch to 2nd Instar (1st instar stage)
The 2021 year shows a relationship of increased average temperatures with a shorter
duration spent in the 1st instar stage (Figure 16), but no statistical meaningful relationship was
found between average increased temperature and 1st instar duration for the 2022 year. In the

44

2021 year, for every degree Fahrenheit increase, the duration spent in 1st instar decreased by
0.37 days. There was an association with increased average relative humidity and a longer
duration spent in 1st instar for both the 2021 and 2022 years. In 2021, for every percent increase
in average relative humidity, the duration spent in 1st instar increased by 0.22 days, and for 2022
for every percent increase in average relative humidity the duration spent in 1st instar increased
by 0.30 days. The median accumulated GDD across the 1st instar stage for 2021 was 175 and
80% of the values were between 140 and 209 GDD (Figure 17). The median accumulated GDD
across the 1st instar stage for 2022 was 198, and 80% of the values were between 104 and 299
GDD.

45

Figure 16. Linear regression output of the average temperature and average relative humidity versus life stage
length for the 1st instar stage. Larvae cups in Raven greenhouse in red, larvae cups for Turtle greenhouse in blue
(n=267 individual larval cups). In 2021, higher average temperatures for 1st instar stage were associated with
shorter life stage duration, but in 2022 no statistical meaningful relationship for these variables was found (for
2021, F1,173 = 72.6, p <0.001, R2 = 0.30. For 2022, F1,86 = 0.60, p <0.4376, R2 = 0.07). For 2021 and 2022, higher
average relative humidity for egg stage was associated with longer life stage duration (for 2021, F1,173 = 84.5, p
<0.0001, R2 = 0.33. For 2022, F1,86 = 6.12, p <0.0154, R2 = 0.06).

Figure 17., Boxplot and histogram for GDD 1st instar life stage. In 2021, accumulated GDD min = 96, median =
175, max = 268. In 2022, accumulated GDD min = 66, median = 198, max = 379.

2nd to 3rd Instar (2nd instar stage)
In 2021 there was an association with increased average temperatures and a shorter
duration spent in the 2nd instar stage (Figure 18), but no statistical meaningful relationship was
found between average increased temperature and 2nd instar duration for the 2022 year. In the

46

2021 year, for every degree Fahrenheit increase, the duration spent in 2nd instar decreased by
0.34 days. There was an association with increased average relative humidity and a longer
duration spent in 2nd instar for the 2021 year, but no association found in 2022. In 2021, for
every percent increase in average relative humidity, the duration spent in 2nd instar increased by
0.26 days. The median accumulated GDD across the 2nd instar stage for 2021 was 22 and 80% of
the values were between 13 and 137 GDD (Figure 19). The median accumulated GDD across the
2nd instar stage for 2022 was 128, and 80% of the values were between 37 and 244 GDD.

Figure 18. Linear regression output of the average temperature and average relative humidity versus life stage
length for the 2nd instar life stage. Larvae cups in Raven greenhouse in red, larvae cups for Turtle greenhouse in
blue (n=267 individual larval cups). In 2021, higher average temperatures for 2nd instar stage were associated with
shorter life stage duration, but in 2022 no statistical meaningful relationship was found (for 2021, F 1,172 = 123.2, p
<0.0001, R2 = 0.42, for 2022, F1,85 = 1.64, p <0.2029, R2 = 0.02). For 2021, increase in percent average relative
humidity was associated with a longer 2nd instar duration, but in 2022 no statistical meaningful relationship was
found (for 2021, F1,172 = 114.7, p <0.0001, R2 = 0.40. For 2022, F1,85 = 2.74, p <0.1016, R2 = 0.03).

47

Figure 19., Boxplot and histogram for GDD 2nd instar life stage. In 2021, accumulated GDD min = 11, median =
22, max = 218. In 2022, accumulated GDD min = 11, median = 128, max = 350.

Hatch to 3rd Instar
In 2021 and 2022 there was an association with increased average temperatures and a
shorter average length of time between 1st to 3rd instar stage (Figure 20). In the 2021 year, for
every degree Fahrenheit increase, the duration from 1st to 3rd instar decreased by 0.9 days. In the
2022 year, for every degree Fahrenheit increase, the duration from 1st to 3rd instar decreased by
1.2 days. Additionally, there was an association with increased average relative humidity and a
longer 1st to 3rd instar duration for both 2021 and 2022 years. In 2021, for every percent increase
in average relative humidity, the duration from 1st to 3rd instar increased by 0.55 days. In
comparison, in the 2022 year for every percent increase in average relative humidity, the length
from 1st to 3rd instar increased by 0.62 days. The median accumulated GDD across the hatch to
3rd instar period for 2021 was 274 and 80% of the values were between 251 and 309 GDD
(Figure 21). The median accumulated GDD across the hatch to 3rd instar period for 2022 was
324, and 80% of the values were between 246 and 401 GDD.

48

Figure 20. Linear regression output of the average temperature and average relative humidity versus life stage
length for 1st instar to 3rd instar duration. Larvae cups in Raven greenhouse in red, larvae cups for Turtle
greenhouse in blue (n=267 individual larval cups). In 2021 and 2022, higher average temperatures from hatch to
3rd instar stage were associated with shorter life stage duration (for 2021, F 1,171 = 296.4, p <0.0001, R2 = 0.64. For
2022, F1,84 = 30.3, p <0.0001, R2 = 0.27). For 2021 and 2022, increase in percent average relative humidity was
associated with longer hatch to 3rd instar duration (for 2021, F1,171 = 254.1, p <0.0001, R2 = 0.60. For 2022, F1,84 =
16.6, p <0.0001, R2 = 0.17).

49

Figure 21. Boxplot and histogram for GDD hatch to 3rd instar life stage. In 2021, accumulated GDD min = 190,
median = 274, max =367. In 2022, accumulated GDD min =197, median 324, max = 506.

Collected to Third Instar
In 2021 and 2022 there was an association with increased average temperatures and a
shorter average length of time between egg to 3rd instar stage (Figure 22). In the 2021 year, for
every degree Fahrenheit increase, the duration from egg to 3rd instar decreased by 2.1 days. In
the 2022 year, for every degree Fahrenheit increase, the duration from egg to 3rd instar decreased
by 3.1 days. Additionally, there was an association with increased average relative humidity and
a longer egg to 3rd instar duration for both 2021 and 2022 years. In 2021, for every percent
increase in average relative humidity, the duration from egg to 3rd instar increased by 1 day. In
comparison, in the 2022 year for every percent increase in average relative humidity, the length
from egg to 3rd instar increased by 1.2 days.

50

Figure 22. Linear regression output of the average temperature and average relative humidity versus life stage
length for the egg to 3rd instar duration. Larvae cups in Raven greenhouse in red, larvae cups for Turtle greenhouse
in blue (n=267 individual larval cups). In 2021 and 2022, higher average temperatures from egg to 3rd instar stage
were associated with shorter life stage duration (For 2021, F 1,173 = 366.3, p <0.0001, R2 = 0.68. For 2022, F1,87 =
46.9, p <0.0001, R2 = 0.36). For 2021 and 2022, increase in percent average relative humidity was associated with
longer egg to 3rd instar duration (for 2021, F1,173 = 350.3, p <0.0001, R2 = 0.67. For 2022, F1,87 = 36.8, p <0.0001,
R2 = 0.30).

Third to Fifth Instar Life Stage
In 2021 and 2022 there was an association with increased average temperatures and a
shorter average length of time between 3rd to diapause (Figure 23). In the 2021 year, for every
degree Fahrenheit increase, the duration from 3rd instar to diapause decreased by 1 day. In the
2022 year, for every degree Fahrenheit increase, the duration from 3rd instar to diapause
decreased by 0.95 days. Additionally, there was an association with increased average relative

51

humidity and a shorter 3rd to diapause duration for 2021, and an association with increased
average relative humidity and longer 3rd instar to diapause duration for 2022. In 2021, for every
percent increase in average relative humidity, the duration from 3rd instar to diapause decreased
by 1.2 days. In comparison, in the 2022 year for every percent increase in average relative
humidity, the length from 3rd instar to diapause increased by 0.64 days. The median accumulated
GDD across the 3rd to 5th instar stage for 2021 was 315 and 80% of the values were between 258
and 370 GDD (Figure 24). The median accumulated GDD across the 3rd to 5th instar stage for
2022 was 389, and 80% of the values were between 333 and 468.

52

Figure 23. Linear regression output of the average temperature and average relative humidity versus life stage
length for the 3rd instar to diapause duration. Larvae cups in Raven greenhouse in red, larvae cups for Turtle
greenhouse in blue (n=745 individual larval cups). In 2021 and 2022, higher average temperatures from 3rd to 5th
instar stage were associated with shorter life stage duration (for 2021, F 1,458 = 1219, p <0.0001, R2 = 0.72. For
2022, F1,283 = 15.7, p <0.0001, R2 = 0.53). For 2021 and 2022, increase in percent average relative humidity was
associated with longer 3rd to 5th instar duration (for 2021, F1,459 = 96.4, p <0.0001, R2 = 0.17. For 2022, F1,283 =
6.8, p <0.0097, R2 = 0.02).

Figure 24. Boxplot and histogram for GDD third to 5th instar life stage. In 2021, accumulated GDD min = 0,
median = 315, max = 428. In 2022, accumulated GDD min = 261, median = 389, max = 701.

Collected to 5th instar (diapause):
In 2021 there was an association with increased average temperatures and a shorter
average length of time between egg to diapause stage, but not for 2022 (Figure 25). In the 2021

53

year, for every degree Fahrenheit increase, the duration from egg to diapause decreased by 3.1
days. Additionally, there was an association of increased average relative humidity and a longer
egg to diapause duration for 2021. In 2021, for every percent increase in average relative
humidity, the duration from egg to diapause increased by 0.56 days. In 2022, the relationship
between egg to diapause duration and average relative humidity was curvilinear (Figure 25).
Shorter durations were associated with both low average RH (<68.5) and high average RH (>70).

Figure 25. Linear regression output of the average temperature and average relative humidity versus life stage
length for the egg to diapause duration. Larvae cups in Raven greenhouse in red, larvae cups for Turtle greenhouse
in blue (n=63 individual larval cups). In 2021, higher average temperatures from egg to diapause instar stage were
associated with shorter life stage duration, but no statistically meaningful relationship was found for 2022 (for
2021, F1,39 = 63.5, p <0.0001, R2 = 0.62. For 2022, F1,19 = 1.4, p <0.2505, R2 = 0.07). For 2021 and 2022, increase
in percent average relative humidity was associated with longer egg to diapause instar duration, for 2022 a
polynomial curve fit this relationship better than a line (for 2021, F1,39 = 6.9, p <0.0001, R2 = 0.12. For 2022, F1,19
= 8.9, p <0.0019, R2 = 0.48).

54

Table 1. Parameter estimates (effect size) for average temperature and average relative humidity by development
period per year. All reported effect sizes with a standard error (SE) are statistically significant (p < 0.05). NS = no
significant relationship.

2021

2022

2021

2022

Life Stage

Avg T (SE)

Avg T (SE)

Avg RH

Avg RH

Egg
1st Instar
2nd Instar
Hatch to 3rd
3rd to 5th
Collected to 3rd
Collected to 5th

-1.06 (0.11)
-0.37 (0.04)
-0.34 (0.03)
-0.9 (0.05)
-1.0 (0.03)
-2.1 (0.11)
-3.1 (0.39)

0.17
0.22
0.26
0.55
-1.20
1.00
0.56

NS
0.3
NS
0.62
0.64
1.2
n/a

-0.59 (0.10)
NS
NS
-1.2 (0.23)
-0.95 (0.23)
-3.1 (0.45)
NS

Table 2. Hatch rate, egg estimates, and larvae counted at 3rd instar by period per year.

Year
2021
2022

Larvae at 3rd instar
6433
3216

Egg estimate
7114
3319

Hatch rate
0.90
0.97

Table 3. Accumulated GDD values by life stage period and quartiles per year.

2021

2022

Life stage

10th %

Median

90th %

10

Median

90

Egg
1st instar
2nd instar
3rd – 5th instar
1st – 3rd instar

220
140
11
0
190

225
175
21
315
274

287
208
218
429
367

147
66
11
261
197

239
198
128
389
324

314
299
350
701
506

55

Discussion
The duration of life stages based on average temperature and relative humidity were
highly variable across different instars stages. All statistically significant periods of development
showed an association with increased average temperature and a shorter amount of time spent in
each instar between the two years. Comparatively, both single instar stages and aggregated instar
stages (such as collected to 5th, hatch to 3rd, etc.) showed an association between a longer time
spent in each life stage and increased relative humidity, apart from the 3rd to 5th duration for the
2021 year. For this period of development, every percent increase in average relative humidity
the time spent in these life stages decreased by 1.2 days (Table 1.). In some years and life stages
there were notable differences in the average temperatures by greenhouse (Raven vs. Turtle).
The Raven greenhouse generally experiences increased temperatures throughout all life stages
due to its placement and smaller infrastructure, which likely explains this difference (Kelli Bush
and Mary Linders, personal communication).
The aggregated instar stages from third to 5th for the 2021 year showed the tightest
association with temperature (Figure 23, R2=0.72) and a shorter length of time spent in this life
stage. For every 1 degree Fahrenheit increase, the total duration spent across this life stage
decreased by 1 day in 2021. Additionally, the collected to 3rd and collected to 5th periods of
development showed the greatest effect sizes (Table 1). For example, in 2021 for the egg to 3rd
instar period, for every degree Fahrenheit increase, the duration of this period of development
decreased by 2.1 days, and in 2022 for every degree increase in temperature, the duration of this
developmental period decreased by 3.1 days.

56

One reason behind why these developmental stages might show such a dramatic decrease
in developmental duration is because they encompass the 4th instar life stages. This individual
life stage was not considered in this study due to a significant amount of data collection errors
between both years, but other studies have observed fourth instar larvae to be more mobile (Haan
et al., 2018), and it is broadly understood that increased temperatures are related to more activity
and movement in checkerspots. The fourth instar stage normally occurs in the summer months
and takes place directly before larvae enter into diapause, so increased temperatures could mean
more ability for the larvae to find and consume food necessary for nutrient uptake, speculated to
be necessary to survive the diapause and post-diapause periods.
Coffee Creek Correctional Facility struggles with higher post-diapause mortality, and
while introducing more UV lighting has found to help alleviate some of this mortality, the
duration that larvae spend within their instar stages is speculated to be another factor in postdiapause death, possibly because larvae—especially in the fourth instar stage—rely on adequate
nutrient uptake to survive through the diapause and wake up periods (Ronda Naseth, personal
communication). Although this study did not place focus on the fourth instar, diapause, or postdiapause stages, the life stages that were analyzed in this research provide an indication that there
is an association with increased temperature and a shorter life stage duration, and this may
influence post-diapause mortality.
In a previous thesis study, Boyd (2021) found that environmental temperature and
humidity targets were not consistently met 100% of the time in past years, with some life stages
for some years meeting environmental target conditions 0% of the time. However, this study also
found that the percentage of success of different life stages, such as the fecundity of butterflies or
larval survival, was not generally associated with whether these environmental targets were met
57

(Boyd, 2021). While meeting environmental targets may not influence metrics such as fecundity,
hatch rate, and larval survival within a captive rearing setting, this study provides support that
environmental factors might play a more significant role out in the field. As temperatures
continue to increase under climate change, based on this captive rearing data it is reasonable to
assume that the duration of larvae life stages will be shortened. This could result in a
phenological mismatch with the butterfly’s host plant.
Based on research conducted by Nate Haan and observations made at Oregon’s Coffee
Creek Correctional Center by butterfly technicians and butterfly husbandry expert Ronda Naseth,
the length of the fourth instar stage in particular may be crucial for larvae post-diapause survival,
and this key life stage occurs during the hotter parts of the year, from June to July (Hann et al.,
2018; Ronda Naseth, personal communication). It is currently unknown as to whether there is an
association with an increase in temperatures and 4th instar duration. Ultimately, a shorter duration
in previous life stages would affect when larvae enter into 4th instar and whether this
developmental period aligns beneficially with plantain abundancy and prior to host plant
senescence. More research in this area and particularly for these later instar stages is needed.
Additionally, more research on temperature effects on checkerspot host plants such as Plantago
lanceolata and Castilleja hispida may help to predict future trends for butterfly and host-plant
phenological changes.
For the life stages and years that showed a statistically significant relationship, increased
relative humidity in this study was weakly associated with longer life stages, except for the 3rd to
5th instar stage for the 2021 year in which the duration of the life stage was decreased by 1.2 days
(Table 1.). Precipitation cues at this stage may be an indication to the organism to prepare for
seasonal dormancy more quickly since fall and winter within its native habitat in the Pacific
58

Northwest tends to experience greater rainfall. It is possible that for the other life stages, more
humid conditions are an indicator to the organism that there would be a greater abundance of a
food source for a longer period of time since food plants would be theoretically obtaining more
water. However, this is speculative—more studies are needed to explore the relationship between
the influence of humidity on the organism.
While accumulated GDD is expected to have an effect on life stage length, for this study,
it was more intuitive to analyze potential associations with average temperature and average
relative humidity and length of life stage, since GDD is typically used in research as an
independent variable when considering the timing of key life stages (hatching, emergence, etc.)
in a species across multiple years. In this study, even though the range of accumulated GDD
across all individual egg cups or larvae cups was generally broad, most values of accumulated
GDD fell within a narrow range. For example, for the 3rd to 5th life stage for the 2021 year, 80%
of accumulated GDD values fell between 258 and 370 GDD (Figure 24). Research by Cayton et
al. (2015), has identified GDD to be used as a better predictor than calendar date when it comes
to assessing different butterfly species and variables such as emergence and abundance across
several years. Cayton et al. (2015) state that for species that respond to GDD, ecological
responses can be predicted under global warming as opposed to just described. Based on that
most values of accumulated GDD fell within a narrow range in this study in addition to Cayton
et al. (2015)’s findings, it is possible that GDD may be used effectively as a variable TCB
phenology in the wild across multiple years.
Ultimately, it is worth noting that this study could benefit from being repeated with the
control for other variables. Food quality and quantity of leaves placed within each larval cup is
not currently recorded or tracked by butterfly technicians. It is unknown whether other variables,
59

such as iridoid glycoside concentrations within the larvae’s food source for example, may be
influencing checkerspot life stage durations within at least some of these results. Iridoid
glycoside concentrations within individual host plants are understood to potentially influence
oviposition selection (Aubrey, 2013), and may also influence the triggering of key events for
other life stages, such as diapause and the percent of post-diapause larvae that return to diapause
rather than pupate.

Limitations
There were several limitations to this research. A portion of the data for the 2022 year
was excluded due to inconsistencies in being recorded, and this was ultimately why the duration
of the 4th instar life stage was not analyzed as a variable in this study. The 4th instar life stage is
particularly tricky for butterfly technicians and coordinators to correctly assess, which is likely
why much of this data went unrecorded. This was also true for a large portion of the 2nd instar
stage for the 2022 year. These larvae sometimes get incorrectly judged as being larger 1st instar
larvae due to technicians being unable to observe visible 2nd instar molts.
Collecting data within a prison environment also has its challenges. Butterfly technicians
are sometimes called inside or are otherwise have limited time in the greenhouse lab spaces,
which could cause technicians to prioritize husbandry care such as quickly feeding larvae and
result in data being not recorded or not recorded accurately. Communication between
Sustainability in Prisons Project coordinators and butterfly technicians is also difficult, since
remote electronic technology for communication such as cell phones disallowed for incarcerated
populations, and messages sent through prison staff are sometimes delayed.

60

Butterfly technicians and coordinators both have to read and internalize specific details
within scientific protocols for the captive rearing of the butterfly, and butterfly technicians
sometimes have to juggle several procedures during the active seasons since life stages overlap.
This sometimes results in details being confused or forgotten. A complete revision of all
procedures was done during the diapause stage of the 2022 year after the data for this thesis was
recorded, with additional details and corrections being implemented into the procedures.
Finally, the 2022 year was especially challenging for the butterfly program. The prison
facility the butterfly program is located in was still under COVID-19 restrictions for most of the
year, transitioning out of these restrictions towards the winter. Butterfly technicians from the two
different units were expected to socially distance and keep separated from one another, with
members of one unit per each greenhouse. These were also the prison’s guidelines for the 2021
year. At the beginning of the 2022 year, all experienced butterfly technicians had transitioned out
of the program, in addition to the butterfly coordinator for this year being new to the job,
resulting in a team more inexperienced than previous years.

Future research
The captive rearing program for Taylor’s Checkerspots could benefit from additional
research. Analyzing how temperature and humidity affect the fourth instar, diapause, and postdiapause instar life stage durations could be used to obtain a better understanding of how the
length of these life stages influence post-diapause mortality rates. Other metrics drawing from
these life stages that are collected by butterfly technicians and coordinators, such as percent of
larvae into diapause and percent of larvae released, could also then be analyzed and compared
with previous studies. The 2021 and 2022 years of the butterfly program did not do captive
breeding of Taylor’s checkerspots as was done in years prior. Captive breeding was restarted
61

during the 2023 year, and so analyzing life stages and variables pertaining to the pupae, eclosion,
and breeding stages could provide insight as to how life stage duration affects butterfly
emergence and copulation success. Furthermore, analyzing more than two years of data could
help paint a broader picture of how temperature and humidity influence larval life stage length
and development.
Finally, studies that focus on the influence of plantain quality, plantain abundance
provided to larvae in cups, and iridoid glycoside concentrations of SPP butterfly program
plantain plants could help butterfly technicians, butterfly coordinators, and program partners
such as WDFW better understand how food quantity and quality impacts larval development and
life stage duration.

62

References
Adedoja, O., Kehinde, T., & Samways, M. J. (2019). Time since fire strongly and variously
influences anthophilous insects in a fire-prone landscape. Ecosphere, 10(9), e02849.
Altermatt, F. (2010). Climatic warming increases voltinism in European butterflies and moths.
Proceedings of the Royal Society B: Biological Sciences, 277(1685), 1281–1287.
Aubrey, D. (2013). Oviposition preference in Taylor's checkerspot butterflies (Euphydryas
editha taylori): Collaborative research and conservation with incarcerated
women (Masters thesis, The Evergreen State College).
Bachelet, D., Johnson, B. R., Bridgham, S. D., Dunn, P. V., Anderson, H. E., & Rogers, B. M.
(2011). Climate change impacts on western pacific northwest prairies and savannas.
Northwest Science, 85(2), 411–429.
Barclay, E., Arnold, M., Anderson, M. J., & Shepherdson, D. (2009). Husbandry manual:
Taylor’s checkerspot (Euphydryas editha taylori). Oregon Zoo, Portland, Oregon, USA.
Bennett, N., Severns, P. M., Parmesan, C., & Singer, M. C. (2015). Geographic mosaics of
phenology, host preference, adult size and microhabitat choice predict butterfly resilience
to climate warming. Oikos, 124(1), 41–53. https://doi.org/10.1111/oik.01490
Bennett, V. J., Pack, S. M., Smith, W. P., & Betts, M. G. (2013). Sex-biased dispersal in a rare
butterfly and the implications for its conservation. Journal of Insect Conservation, 17(5),
949–958.
Bonhomme, R. (2000). Bases and limits to using ‘degree. Day’units. European Journal of
Agronomy, 13(1), 1–10.

63

Boyd, C. (2021). Importance of accurate environmental conditions during captive rearing of an
endangered butterfly (Euphydryas editha taylori): Collaborative research with
Sustainability in Prisons Project and incarcerated technicians [Masters thesis, The
Evergreen State College]

Boyd, R. T., & Lake, F. K. (2021). Indians, fire, and the land in the Pacific Northwest. Oregon
State University Press.

Bristow, L. V., Grundel, R., Dzurisin, J. D., Wu, G. C., Li, Y., Hildreth, A., & Hellmann, J. J.
(2022). Warming experiments test the sensitivity of an endangered butterfly across life
history stages.
Buckingham, D. A., Linders, M., Landa, C., Mullen, L., & LeRoy, C. (2016). Oviposition
preference of endangered taylor’s checkerspot butterflies (Euphydryas editha taylori)
using native and non-native hosts. Northwest Science, 90(4), 491–497.
Bush, K. (2021). Diapause. The Sustainability in Prisons Project.
Bush, K. (2022a). Mission Creek Procedures: Oviposition. The Sustainability in Prisons Project.
Bush, K. (2022b). Eggs to Third Instar Larvae Procedures 2022. The Sustainability in Prisons
Project.
Cayton, H. L., Haddad, N. M., Gross, K., Diamond, S. E., & Ries, L. (2015). Do growing degree
days predict phenology across butterfly species? Ecology, 96(6), 1473–1479.
Crawford, R. C., & Hall, H. (1997). Changes in the south Puget prairie landscape. Ecology and
conservation of the south Puget sound prairie landscape. The Nature Conservancy,
Seattle, WA, 11-15.
64

Curry, K. (2019). Reproduction in captivity of the endangered Taylor's checkerspot butterfly
(Euphydryas editha taylori): A collaboration with the Sustainability in Prisons Project.
[Masters thesis, The Evergreen State College]
D’Antonio, C. M., & Vitousek, P. M. (1992). Biological invasions by exotic grasses, the
grass/fire cycle, and global change. Annual Review of Ecology and Systematics, 23(1),
63–87.
Dunwiddie, P. W., Haan, N. L., Linders, M., Bakker, J. D., Fimbel, C., & Thomas, T. B. (2016).
Intertwined fates: Opportunities and challenges in the linked recovery of two rare species.
Natural Areas Journal, 36(2), 207–215.
Ehrlich, P. R., & Hanski, I. (2004). On the wings of checkerspots: A model system for population
biology. Oxford University Press.
Ehrlich, P. R., Murphy, D. D., Singer, M. C., Sherwood, C. B., White, R. R., & Brown, I. L.
(1980). Extinction, reduction, stability and increase: The responses of checkerspot
butterfly (Euphydryas) populations to the California drought. Oecologia, 46(1), 101–105.
Forrest, J. R. (2016). Complex responses of insect phenology to climate change. Current Opinion
in Insect Science, 17, 49–54.
Ghazanfar, M., Malik, M. F., Hussain, M., Iqbal, R., & Younas, M. (2016). Butterflies and their
contribution in ecosystem: A review. Journal of Entomology and Zoology Studies, 4(2),
115–118.

65

Grosboll, D. N. (2011). Taylor's checkerspot (Euphydras editha taylori) oviposition habitat
selection and larval hostplant use in Washington State (Masters thesis, Evergreen State
College).
Haan, N. L., Bakker, J. D., Dunwiddie, P. W., & Linders, M. J. (2018). Instar‐specific effects of
host plants on survival of endangered butterfly larvae. Ecological Entomology, 43(6),
742-753.
Halofsky, J. E., Peterson, D. L., & Harvey, B. J. (2020). Changing wildfire, changing forests:
The effects of climate change on fire regimes and vegetation in the Pacific Northwest,
USA. Fire Ecology, 16(1), 1–26.
Hellegers, M., van Swaay, C. A., van Hinsberg, A., Huijbregts, M. A., & Schipper, A. M. (2022).
Modulating effects of landscape characteristics on responses to warming differ among
butterfly species. Frontiers in Ecology and Evolution, 10, 873366.
Hill, K., Kronland, B., & Martin, A. (2017). Fire effects monitoring Joint Base Lewis-McChord
annual report. Center for Natural Lands Management Olympia.
Hill, K., & Martin, A. (2019). Fire Effects Monitoring Joint Base Lewis-McChord Annual
Report. Center for Natural Lands Management, 2.
Janzen, D. H. (1980). When is it coevolution?
LeRoy, C. J., Bush, K., Trivett, J., & Gallagher, B. (2012). The Sustainability in Prisons Project:
An overview (2004-2012). The Evergreen State College & Washington State Department
of Transportation. 978-0-9886415-0-1

66

Luoto, M., Heikkinen, R. K., Pöyry, J., & Saarinen, K. (2006). Determinants of the
biogeographical distribution of butterflies in boreal regions. Journal of Biogeography,
33(10), 1764–1778.
Martin, A., & Kronland, B. (2015). Prairie habitat management Joint Base Lewis-Mccord 2014
annual report. Center for Natural Lands Management.
Menke, C., & Kaye, T. (2016). Action Plan for Taylor’s Checkerspot Butterfly Conservation in
Oregon. Institute for Applied Ecology.
Miller, P., Lanier, W., & Brandt, S. (2001). Using growing degree days to predict plant stages.
Ag/Extension Communications Coordinator, Communications Services, Montana State
University-Bozeman, Bozeman, MO, 59717(406), 994–2721.
Murphy, D. D., & Weiss, S. B. (1988). Ecological studies and the conservation of the bay
checkerspot butterfly, Euphydryas editha bayensis. Biological Conservation, 46(3), 183–
200.
Oregon Zoo. (2019, March 29). With inmates’ help, rare NW butterfly is homeward bound.
Oregon Zoo. https://www.oregonzoo.org/news/2019/03/inmates-help-rare-nw-butterflyhomeward-bound
Parmesan, C., Williams-Anderson, A., Moskwik, M., Mikheyev, A. S., & Singer, M. C. (2015).
Endangered Quino checkerspot butterfly and climate change: Short-term success but
long-term vulnerability? Journal of Insect Conservation, 19(2), 185–204.
Parry, M. L., & Carter, T. R. (1985). The effect of climatic variations on agricultural risk.
Climatic Change, 7(1), 95–110.

67

Potter, A. E. (2016). Periodic status review for taylor’s checkerspot [Review]. Washington
Department of Fish and Wildlife.
Potts, S. G., Vulliamy, B., Dafni, A., Ne’eman, G., O’Toole, C., Roberts, S., & Willmer, P.
(2003). Response of plant-pollinator communities to fire: Changes in diversity,
abundance and floral reward structure. Oikos, 101(1), 103–112.
Poulos, L. P., & Roy, B. A. (2015). Fire and false brome: How do prescribed fire and invasive
Brachypodium sylvaticum affect each other? Invasive Plant Science and Management,
8(2), 122–130.
Reed, P. B., Pfeifer-Meister, L. E., Roy, B. A., Johnson, B. R., Bailes, G. T., Nelson, A. A.,
Boulay, M. C., Hamman, S. T., & Bridgham, S. D. (2019). Prairie plant phenology driven
more by temperature than moisture in climate manipulations across a latitudinal gradient
in the Pacific Northwest, USA. Ecology and Evolution, 9(6), 3637–3650.
Renner, S. S., & Zohner, C. M. (2018). Climate change and phenological mismatch in trophic
interactions among plants, insects, and vertebrates. Annual Review of Ecology, Evolution,
and Systematics, 49(1), 165–182.
Richards, K. D. (1981). In search of the Pacific Northwest: The historiography of Oregon and
Washington. The Pacific Historical Review, 415–443.
Roy, D. B., Oliver, T. H., Botham, M. S., Beckmann, B., Brereton, T., Dennis, R. L., Harrower,
C., Phillimore, A. B., & Thomas, J. A. (2015). Similarities in butterfly emergence dates
among populations suggest local adaptation to climate. Global Change Biology, 21(9),
3313–3322.

68

Severns, P. M. (2008). Interactions between two endangered butterflies and invasive, exotic
grasses in western Oregon, USA. Endangered Species Update, 25(2).
Severns, P. M., & Grosboll, D. (2011). Patterns of reproduction in four Washington State
populations of Taylor’s checkerspot (Euphydryas editha taylori) during the spring of
2010. The Nature Conservancy, Olympia, WA.
Severns, P. M., & Warren, A. D. (2008). Selectively eliminating and conserving exotic plants to
save an endangered butterfly from local extinction. Animal Conservation, 11(6), 476–
483.
Singer, M. C., & Parmesan, C. (2010). Phenological asynchrony between herbivorous insects
and their hosts: Signal of climate change or pre-existing adaptive strategy? Philosophical
Transactions of the Royal Society B: Biological Sciences, 365(1555), 3161–3176.
Schultz, C. B., Henry, E., Carleton, A., Hicks, T., Thomas, R., Potter, A., ... & Reader, B. (2011).
Conservation of prairie-oak butterflies in Oregon, Washington, and British
Columbia. Northwest Science, 85(2), 361-388.
Stinson, D. W. (2005). Draft Washington State Status Report for the Mazama pocket gopher,
streaked horned lark, and Taylor’s checkerspot. Washington Department of Fish and
Wildlife, Olympia, 138.
USGCRP, 2018: Impacts, Risks, and Adaptation in the United States: Fourth National Climate
Assessment, Volume II [Reidmiller, D.R., C.W. Avery, D.R. Easterling, K.E. Kunkel,
K.L.M. Lewis, T.K. Maycock, and B.C. Stewart (eds.)]. U.S. Global Change Research
Program, Washington, DC, USA, 1515 pp. doi: 10.7930/NCA4.2018.

69

Miller, H. & Reese, M. (n.d.) A curriculum project for Washington State schools. Indians and
Europeans on the Northwest Coast, 1774-1812. University of Washington, Department of
History. https://sites.uw.edu/cspn/indians-and-europeans/
Vickery, M. (2008). Butterflies as indicators of climate change. Science Progress, 91(2), 193–
201.
Westerling, A. L. (2016). Correction to ‘Increasing western US forest wildfire activity:
Sensitivity to changes in the timing of spring.’ Philosophical Transactions of the Royal
Society B: Biological Sciences, 371(1707), 20160373.
Williams, D. B. (2016, December 13). Modern plant communities in the Puget lowland begin to
thrive around 7,000 years ago. Historylink.org. Retrieved April 16, 2023, from
https://historylink.org/File/20236

70



Item sets

The Evergreen State College Masters Theses: Environmental Studies

Media

Thesis_MES_2023_DunningR.pdf