K , . .
HELEJI

3
«mt,

r.

3 , a. ‘u
mhvum
I; .

my

 

 

!
E. v a
9.5...

‘1 hi

i~
‘ I 'v. . a
Kuvrwﬁan
:11»?

.‘ﬂm

 

 

by.
LIV...

u A - l .. - I I.

:.;_u.:¢v!q€1»;. W .1 It... .3...

ll

THESIS

Date

Millillillllilllililllillllllilllllllill

3 1293 0168

This is to certify that the

thesis entitled

RESPONSE OF WET MEADOW TUNDRA TO INTERANNUAL
AND MANIPULATED TEMPERATURE VARIATION:
IMPLICATIONS FOR CLIMATE CHANGE RESEARCH

presented by

Robert D. Hollister

has been accepted towards fulﬁllment
of the requirements for

MasceIL degree in W Plant Pathology

pmw who

<41ajor professor

 

5/11/98

MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

LIBRARY

A Michigan State
University

 

 

 

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINE-3 return on or before date due.

 

DATE DUE MTE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/98 mus-n14

 

RESPONSE OF WET MEADOW TUNDRA TO
INTERANNUAL AND MANIPULATED TEMPERATURE VARIATION:
IMPLICATIONS FOR CLIMATE CHANGE RESEARCH

By

Robert D. Hollister

A THESIS
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Botany and Plant Pathology

1998

ABSTRACT

RESPONSE OF WET MEADOW TUNDRA TO
INTERANNUAL AND MANIPULATED TEMPERATURE VARIATION:
IMPLICATIONS FOR CLIMATE CHANGE RESEARCH

By

Robert D. Hollister

This research is a contribution to the lntemational Tundra Experiment (ITEX).
ITEX was established to monitor and make realistic predictions of plant response to
climate change. The hypothesis is that short-term warming of ambient temperature will
lead to accelerated phenology and increased vigor. Measured characters were date of
ﬂowering, number of ﬂowers, stature, number of leaves, and leaf length. Twenty-four
small open-top chambers were used to passively warm canopy temperatures in wet
meadow tundra at Barrow, Alaska during the summers of 1995 and 1996. Fortuitously
the seasonal average temperature difference due to chamber warming and interannual
variability were both approximately 1.5 °C; this allowed comparisons of species response
to warming caused by the two mechanisms. The statistical signiﬁcance of species
responses to chamber warming and interannual warming were Similar 70% of the time.
All Species showed signiﬁcant trends of increased vigor or earlier phenologic
development under warmer canopy temperatures for at least one character. The most

consistent plant response to warmer temperature was increased plant stature.

ACKNOWLEDGEMENTS

I would ﬁrst like to thank my advisor Patrick J. Webber for sharing his knowledge
and the wonderful opportunities he has given me. I am indebted to Dr. Christian Bay for
introducing me to Arctic research and teaching me the species of the North Slope and Dr.
Frederick E. Nelson for providing every opportunity he could for me to do research in
Barrow. I could not have completed this undertaking without the long hours and hard
work provided by the people I worked with and I thank them all including Stephanie
Grimes, Lisa Walker, Brian Noyle, William Chu, Ian Ramjohn, Lisa Koch, Mark
Peterson, Laura Stack, Sharon Osborn, Anna Noson, David Conlin, Bennett Weinstein,
Matthew Guisbert, and MaryGrace Villanueva. I am grateful to all the researchers who
shared their experience and friendship. Dr. Dave Norton, Dr. Kenneth I-Iinkel, Dr. Jerry
Brown, Nikolay Shiklomanov, Jerry Mueller, Anna Klene, Dr. Ronald Paetzold, Dr. Jim
Bockheim, Dr. Wendy Eisner, Dr. Samuel Outcalt, Lynn Everett, and Laura Miller all
were generous. I thank the Barrow community particularly Dario Leiba and the people at
the Barrow Arctic Science Consortium (BASC), most notably Dr. Glenn Sheehan and
Richard Glenn. I am indebted to Malcolm Gaylord and Dan Enders at the National
Oceanic and Atmospheric Administration (N 0AA) for data, knowledge, friendship, and
shelter. I thank the International Tundra Experiment (ITEX) community for all their

friendly advice and counsel particularly Dr. Ulf Molau. I am indebted to Professors Peter

iii

G. Murphy and Thomas M. Burton for their help in getting me established in science and
continual conversations and advice as committee members and friends. I owe my
statistical knowledge to Professor Oliver Schabenberger. Funding for this research was
provided by the Ofﬁce of Polar Programs (OPP) of the National Science Foundation

(N SF) (Grants no. OPP-9318528 and OPP-9714103) with logistics support provided by
the Polar Ice Coring Ofﬁce in Lincoln. Nebraska. I thank the people at both
organizations for their support particularly Dr. Michael Ledbetter of NSF. Finally I

would like to thank my family and friends for continual support.

iv

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... viii

LIST OF FIGURES ............................................................................................................ x
Chapter I

INTRODUCTION .............................................................................................................. l

1.1 RATIONALE ..................................................................................................... I

1.2 BACKGROUND ............................................................................................ .2

1.2 1 Global Climate Change 2

1211 GlobalChmateModels .................................................... 3

1.2.1 -2 Current Evidence for Warming ....................................... 5

1.2.2 Vegetation and Climate Change Relations ..................................... 6

1.2.2-l Carbon Dioxide and Plants ............................................... 6

1.2.2-2 Climate and Plants ........................................................... 7

1.2.3 The Arctic System ...................................................................... .10

1.2.3-l Arctic Plant Adaptations .............................................. .11

1.2.3-2 Barrow Alaska .............................................................. .14

1.3 RESEARCH FRAMEWORK ............................................................................. .15

131 Arctic System Science ................................................................. .15

13 .2 The Arctic Ecology Lab. 16

13 3 The International Tundra Experiment ......................................... .17

13 3-1 The Network and its History ........................................ .19

1.3.3-2 The Chamber 21
I..4 1 Phenology Response22
I..4 2 Vigor Response 23
I..4 3 Species Patterns of Response 24

Chapter 11
METHODS ........................................................................................................... .25
11.1 STUDY AREA ................................................................................................ 25
11.1.1 Site Establishment .................................................................... .28
11.2 ABIOTIC DOCUMENTATION AND CHAMBER PERFORMANCE ....................... .31
11.2.1 Chamber Design .......................................................................... .31
11.2.2 Climate Monitoring .................................................................... .31
11.2.2-1 Macroclimate Monitoring ............................................ .31

V

11.2.2-2 Microclimate Monitoring .............................................. 31
11.2.2-2.1 Horizontal and Vertical

 

 

Temperature Distribution .................... 33

11.2.2-2.2 Light Distribution - - -- -.......35

11.2.3 Below Ground Monitoring - _ 36
11.2.3-l Soil Temperature ......................................................... .36

11.2.3-2 Active Layer Monitoring ............................................. .37

11.2.4 Degree Day Calculation .............................................................. .38

11.3 Vascular Plant Monitoring ........................................................................... .39

11.3.1 Phenologic Monitoring. . . . . . .. .. . . . ...39
11.3.2 Vigor Monitoring 39
11.4 Data Analysis ............................................................................................. .44
11.4.1 Active LayerComparisons 44

Chapter 111
ABIOTIC DOCUMENTATION ANDCHAMBER PERFORMANCE 48
1111 CLIMATICDATA ............................................................................... 48
111.2 CHAMBER PERFORMANCE .......................................................... 49
111.2.1 Temperature and Humidity ......................................................... 49
[11.2.2 Vertical and Horizontal Patterns of Chamber Warming .. ........... 51
111.2.3 Light Distribution within a Chamber .......................................... 51
[113 BELOW GROUND RESPONSES TO TEMPERATURE. . . . . . . . . . . 55
1112.] Soil Temperature ......................................................................... 55
111.2.2 Active Layer Thickness . . ........................................................... 57
111.4 ACCUMULATION OF THAWING DEGREE DAYS . . ......................................... 53
111.5 DISCUSSION AND CONCLUSIONS ....................................................... 65
Chapter IV
PLANT RESPONSES TO TEMPERATURE ......................................................................... 65
NJ RESPONSE OF FIDWERING PHENOLOGY ................................................... 65
IV.1.1 Hypothesis 1a: Flowering will occur earlier under
warmer conditions. ............................................ 65
IV.1.2 Hypothesis lb: Flowering will occur at approximately the
same degree day threshold. ............................... 66
IV.2 VIGOR RESPONSE 69
IV .2.1 Hypothesis 2a: Plants will produce more ﬂowers under
warmer conditions. . . . ........................................ 69
IV.2.2 Hypothesis 2b: The height of inﬂorescence will be taller
under warmer conditions. .................................. 71
IV .2.3 Hypothesis 2c: The length of leaves will be longer under
warmer conditions. ............................................ 73

1V3 SPECIES PATTERNS OFRESPONSE 76

vi

IV.3.1 Hypothesis 3a: Short-term plant phenologic response to

canopy warming will mirror responses

observed in warm years. ................................. .76
N32 Hypothesis 3b: Short-term response of plant vigor observed

in the chambers will mirror responses of

controls observed in warm years. ..................... 76

IV.3.3 Hypothesis 4: Plant species will respond individualistically to
temperature. ....................................................... 78
NA DISCUSSION AND CONCLUSIONS ................................................................ .79
IV .4.1 Hypothesis 1: Response of Flowering Phenology .................... .79
IV.4.2 Hypothesis 2: Vigor Response ................................................ 81
IV .4.3 Hypothesis 3: Chamber Efﬁcacy .. . . . ......................................... 85
IV.4.4 Hypothesis 4: Individualistic Nature of Species ......................... 89
IV.4.5 Summary and Conclusions ........................................... 90

Chapter V

CONCLUDINGREMARKS 93
V.1 RELEVANCE OF THE STUDY m RELATION To CLIMATE CHANGE ................... 93
v.2 SYNTHESIS OF THE SYSTEM RESPONSE ......................................................... 94
v.3 IMPLICATIONS OF THE STUDY ........................................................................ 96
v.4 FUTURE WORK . . ...................................................................................... 97
APPENDIX . .............................................................................................................. 101
LITERATURE CITED ...................................................................................................... 1 l3

vii

Table 11-1
Table 11-2
Table II-3
Table 11-4

Table 111-1

Table IV-l

Table IV -2

Table IV-3

Table IV-4

Table IV-S

Table IV-6

LIST OF TABLES

Climate data for the Barrow region (Brown et al. 1980).
Species composition of the ﬁeld plots.

Phenologic measures recorded for all species monitored.
Vigor measures recorded for all species monitored.

Summer average, maximum, and minimum daily temperature
and relative humidity of chambers and controls and
climatological data of the nearby NOAA station from snowmelt
until August 18‘“ for 1995 and 1996.

Percentage of monitored individuals that ﬂowered during the
two years in both treatments.

Statistical signiﬁcance reported from post factor tests of the
population means of the 1995 OTC compared against the 1996
control. The group that responded earlier on average is also
presented.

Summary table of signiﬁcance of year and treatment of all
Species. All responses were positive in relation to warmth (i.e.,
earlier ﬂowering; greater numbers of ﬂowers, length of
inﬂorescence, length of leaf, and vegetative measure) except for
E. triste which had a decreased vegetative response in the
warmer year (bold).

Summary table of P-values from overall tests for all measures
presented.

Summary table of the number of species with signiﬁcant
response in their phenology of ﬂower opening (number
responding I number examined.

Summary table of the number of species with signiﬁcant
responses in height of inﬂorescence and number of ﬂowers
(number responding I number examined).

viii

27

28

42

43

49

71

77

78

79

82

Table IV-7

Table V-l

Table A-1

Table A-2

Summary table of the number of species that showed signiﬁcant
vegetative response (number responding / number examined).

Three potential small scale experiments speciﬁcally designed to
test hypotheses presented within this thesis.

Average, maximum, and minimum daily temperature and
relative humidity of chambers and controls and climatological
data of the near by N 0AA station from snow melt until August
18" for the years 1995 and 1996

Average active layer thickness and standard deviation for the

ﬁeld seasons 1995 and 1996 (N for OTC = 24; N for control =
96).

ix

86

98

107

111

Figure 1-1.

Figure 1—2

Figure 1-3

Figure 1-4

Figure 1-5

Figure 11-1

Figure 11-2

Figure 11-3

Figure 114
Figure 11-5

Figure 11-6

LIST OF FIGURES

Conceptual diagrams of a generalized GCM. (A) The basic

processes representing equations of a model (Dickson 1987). (B)
The scale at which these equations are run (Henderson-Sellers

1990).

Diagram of mean, maximum and minimum temperatures, snow
depth, active layer thickness, and solar radiation at Barrow (from
Chapin & Shaver 1985a).

Logos of the research efforts with which this project is
networked: (A) the Arctic Ecology Lab, (B) Arctic System
Science and (B) the International Tundra Experiment.

Map of the ITEX ﬁeld sites. Site 13 is Barrow (Molau &
Malgaard 1996).

An open-top chamber (OTC).

Map of Alaska and the Barrow Pennisula showing the ITEX ﬁeld
sites and other historical research locations.

Map of the research area, showing the ARCSS lxlkm grid
(crosses), the ﬁeld sites, beach ridge, and other prominent
features. The Boundary of the Barrow Environmental
Observatory (BBC) is shown on inset (Hinkel et at. 1996).

The ﬁeld site shortly after establishment in 1995 (courtesy of
Christian Bay).

Map of ﬁeld site showing each plot.
The design of the open-top chambers.

The spatial arrangement of temperature probes: (A) horizontal
disu'ibution and (B) vertical distribution.

11

18

2O

21

26

26

29

30

32

34

Figure 11-7

Figure 11-8

Figure n-9

Figure 11-10

Figure 111-1

Figure III-2

Figure 1H-3

Figure 111-4

Figure m-s

Figure III—6

Figure III-7

Figure IV-l

The placement of four light sensors within a chamber.

The chamber within which soil temperature monitoring was
conducted.

Probing the active layer.

Representation of the plant monitoring scheme: (A) recording
sheet, (B) marked individual, and (C) map of plot.

The course of mean temperature and relative humidity in OTCS
and control plots for three different types of days: a sunny day
(8/8/95); a typical day (8/6/1995); and a cloudy day (7/30/95) (N
for OTC =18; N for control = 18).

The mean (thick line) and range (thin line) course of temperature
and relative humidity in OTCS and control plots on a sunny day
(8/8/95) (N for OTC = 18; N for control = 18).

The vertical and horizontal temperature distribution of the
average daily mean, maximum and minimum within a chamber
from June 20 - July 28, 1995 (N = 1).

Light distribution relative to natural conditions within a chamber
from August 1 — 21, 1996 (N=1 and N=2 for the center).

Soil warming under the chambers in two community types and
their respective controls. Data collected were from the same time
period (July 10 - August 18, 1996) and the two chambers were
within 100 meters of each other (N = l).

Progression of active layer throughout the 1995 and 1996 ﬁeld
seasons (N for OTC = 24; N for control = 96).

Mean (thick line) and range (thin line) degree day accumulation
from day of snow melt for the ﬁeld seasons 1995 and 1996 (N 2
7).

Julian date of ﬂowering for all species. Histogram bars represent
averages, with standard error of the mean as error bars. P-values
of less than 0.05 reported from a two way ANOVA and samples
Size are given for each species. Overall signiﬁcance (top left) is
reported ﬂour a three way ANOVA run on all forbs and

grarninoids.

xi

35

36

37

50

52

53

54

56

57

59

67

Figure 1V-2

Figure 1V-3

Figure lV-4

Figure IV-S

Figure 1V-6

Figure V-l

Figure A-l

Degree day of ﬂowering for all species. Histogram bars
represent averages, with standard error of the mean as error bars.
P values less than 0.05 reported from a two way AN OVA and
samples size are given for each species. Overall signiﬁcance (top
left) is reported from a three way ANOVA run on all forbs and

graminoids.

Number of individuals ﬂowering within a plot. Histogram bars
represent averages, with standard error of the mean as error bars.
P-values of less than 0.05 reported from a two way ANOVA and
samples size are given for each species. Overall signiﬁcance (top
left) is reported from a three way ANOVA run on all forbs and
graminoids.

End of season measures of the length of inﬂorescence.
Histogram bars represent averages, with standard error of the
mean as error bars. P-values of less than 0.05 reported from a
two way AN OVA and samples size are given for each species.
Overall signiﬁcance (top left) is reported from a three way
AN OVA run on all forbs and graminoids.

End of season measures of the length of longest leaf. For species
D. lactea and S. foliolosa measures are the diameter of rosette.
Histogram bars represent averages, with standard error of the
mean as error bars. P-valnes of less than 0.05 reported from a
two way AN OVA and samples size are given for each species.
Overall signiﬁcance (top left) is reported from a three way
ANOVA run on all forbs and graminoids.

End of season measures of vegetative growth (length of the
longest leaf multiplied by the number of leaves). Histogram bars
represent averages, with standard error of th emean as error bars.
P-values of less than 0.05 reported from a two way ANOVA and
samples size are given for each species. Overall signiﬁcance (top
left) is reported from a three way ANOVA run on all forbs and

graminoids.

(A) A conceptua diagram of the tundra plant-soil system
components predicted to respond to temperature change and
monitored by this ongoing project. The principal domains of the
Question Themes (QT) and their regions of overlap indicate
important system linkages. (B) The diagrams in A are extended
to depict the spatial and temporal dimensions of the project.

Speciﬁcation of the StowAwayW temperature data logger
manufactured by Onset Inc.

xii

68

70

72

74

75

102

Figure A-2 Speciﬁcation of the Hobo® temperature data logger 103

manufactured by Onset Inc.

Figure A-3 Speciﬁcation of the StowAwayT“ relative humidity data logger 104
manufactured by Onset Inc.

Figure A4 Speciﬁcation of the Hobo® relative humidity data logger 105
manufactured by Onset Inc.

Figure A-5 Speciﬁcation of the StowAwayTM light data logger manufactured 106
by Onset Inc.

Chapter I
INTRODUCTION

1.1 RATIONALE

In recent years there has been considerable interest in species response to elevated
temperatme and carbon dioxide (C02) levels because of concern over anthroprogenically
enhanced climate change. Various scenarios of vegetation response to warming have
been presented (Bonan et al. 1990, Woodward 1993, Monserud et al. 1993, Prentice &
Sykes 1995, Watson et al. 1996, Shugart & Smith 1996). The magnitude of the potential
inﬂuence climate change could have on the biota was illustrated as early as 1985 by
Emanuel et al. This model used the Holdridge life zones to delineate current biome
boundaries and projected future biome boundaries based on predictions from a global
climate model. Although the model was simplistic and, for reasons presented later
(section 1.2.2-2), likely to be unrealistic, it was useful to exemplify the alarming potential
of future climates to modify present ecosystems including the entire elimination of the
tundra biome under the future global change scenario.

In all global climate change predictions concerning warming due to atmospheric
enrichment of C02, the polar latitudes are projected to warm by a larger amount than
lower latitudes (Houghton et al. 1996). Since polar organisms are adapted to cold
climates, it is of interest to ask how well they can adapt to a warmer climate (Stonehouse
1989, McGraw & Fetcher 1992). This project attempts to determine the Short-term
adjustments in phenology and vigor of plants in a wet meadow tundra at Barrow, Alaska.
It is a contribution to a collaborative project known as the International Tundra

Experiment (ITEX).

1.2 BACKGROUND

1.2.1 Global Climate Change

Climate has never been and will never be static (MacCracken er al. 1990). It is a
dynamic and continually changing interconnection of processes operating at many
different scales of size, magnitude and time. Throughout the history of the Earth the
climate has been changing due to many factors including biotic inﬂuence (Budyko et al.
1987). The addition of oxygen to the atmosphere during the evolution of photosynthesis
greatly inﬂuenced many processes on Earth. Even today the biota (primarily soil fauna)
emits more C02 than humans (Schimel et al. 1995). Currently there is considerable
emphasis on the potential inﬂuence of anthroprogenically increased CO2 levels in the
atmosphere on the Earth’s climate and biota. Atmospheric CO2 levels have been
dynamic throughout the Earth’s history and these changes Show a strong positive
correlation with temperature (MacCraken et al. 1990).

At the turn of the century, Arrhenius predicted that the amount of C02 being
emitted by anthroprogenic activity, such as the burning of coal, could result in changes in
the Earth’s energy balance and likely cause wamring (Arrhenius 1896). This assumption
was based on the simple knowledge that CO2 is an absorber of heat energy and is what is
now termed a greenhouse gas. Climatologists are currently attempting to model the
response of global climate by the use of complex models run on supercomputers, and
these equations are fundamentally based on the same basic assumptions of Arrhenius

(1896).

1.2.1-1 Global Climate Models

Global Circulation Models (GCMs) are a collection of relatively simple equations
attempting to simulate a complex phenomenon (Figure I-2). Most GCMs fundamentally
consist of a relatively small number of equations representing the overarching processes
occurring in the atmosphere (Henderson-Sellers 1990, Houghton 1997). The models run
from half hour to daily time scales on three dimensional atmosphere cells of various sizes
that cover the globe (Houghton et al. 1996, Figure 1-2). The complexity of the models is
primarily due the number of iterations run simultaneously for each representative cell and
the connectiveness of each cell to its neighboring cells (Figure I-l).

Although the GCMs are not in complete agreement, there are some points upon
which nearly all climatologists do agree. Among them, the most relevant to this thesis is
that the globe will warm due to anthroprogenic greenhouse gas emissions such as C02,
methane, and chloroﬂuorocarbons (CFC’s), even in light of all the potential positive and
negative feedback processes identiﬁed (Houghton et al. 1996). Scientists also agree that
this warming will not be evenly distributed in time or space (Houghton et al. 1990). It is
generally agreed that temporal variation of temperature will be greater, causing more
extreme weather events and increased variability (Karl et al. 1995) and that the high
latitudes will warm substantially more than lower latitudes (Weller 1984, Maxwell 1992).
GCMs also predict that there will be higher precipitation levels globally and that these
levels will be unevenly distributed with greater increases near the coastal and montane
regions and lower increases in the centers of the continents. Beyond these predictions,
precipitation models are less agreed on than are temperature models (Rizzo & Wiken

1992, Kattenberg et al. 1996).

m

 

 

 

 

more mono. norm
M
W mm
mm
PM
um

 

 

 

 

// 1W 5;" 7//.

ATMMAOI
Gm
mmmm
held-MW“.

 

Tuneup-30m
Odom-313’

Figure 1-1. Conceptual diagrams of a generalized GCM. (A) The basic processes
representing equations of a model (Dickson 1986). (B) The scale at which these
equations are run (Henderson-Sellers 1990).

1.2.1-2 Current Evidence for Warming

There is evidence that recent global warming is due to anthropogenic activities.
The Intergovernmental Panel on Climate Change (Houghton et al. 1996) clearly states
“the balance of evidence suggests that there is a discernable human inﬂuence on global
climate.” Long-term climatic records show a warming trend over the last century of 1°C.
Many researchers have attributed this warming trend to unrepresentative heat islands
created by the expansion of the cities from which the long-term data are gathered.
Although this may be true in many cases, there are additional independent data sets
which also show warming trends; these include oceans surface temperatures (Cane et al.
1997), permafrost temperatures (Lachenbruch & Marshall 1986), and glacier retreats
(Oerlemans 1994, Dyurgerov & Meier 1997). There is also evidence that the globe is
already experiencing increased variations in temperature and precipitation (Hansen et al.
1998).

Work by Chapman and Walsh (1993) has shown that high latitudes have warmed
more than lower latitudes. They found arctic temperatures have been warming at a rate
of 075°C per decade, although warming was not uniform around the Arctic and in some
areas there was cooling. The Alaskan Arctic is in a region of warming. This warming is
attributed to a multitude of factors but the two most important are: positive feedback from
decreased albedo due to snow melt; and increased heat ﬂow, because nearly half of the

energy in the Arctic is attributed to a ﬂow of heat from lower latitudes (Maxwell 1992).

1.2.2 Vegetaﬁon and Climate Change Relations
1.2.2-1 Carbon Dioxide and Plants

Increased carbon dioxide (CO2) levels can themselves directly affect a plant
community. As a plant opens its stomata to obtain C02, one of the essential building
blocks for photosynthesis, it loses the other, water, via the same stomata. This causes a
perpetual balance between the need to open the stomata to obtain C02 and the need to
close them to prevent desiccation. The ratio of water used: sugar synthesized is known as
the Water Use Efﬁciency (WUE). Many physiological processes alter the WUE; it is
logically predicted that as atmospheric C02 concentrations increase a plant’s WUE
should also increase. This reasoning has led many to predict that plants will increase
their vigor in an atmosphere with increased C02 concentrations (Long 1991, Dahlman
1993, Graves & Reavey 1996). This is especially true in crop production when nutrient
limitations and competition are potentially minimized. The response of native vegetation
and communities is much less predictable because there are many other potential
limitations on plant productivity. Furthermore it is expected that species with different
physiological adaptations and biochemical pathways will respond differently. For
example, C3 plants are expected to respond more than C. plants (Pearcy & Bjorkrnan
1983).

In a review of studies conducted on natural vegetation, Bazzaz (1990) found that
species respond individually to increased C02 and that many factors, particularly
available nutrients and temperature, affect this response. One of the ﬁrst and largest
community-scale C02 enhancement experiments, done on Alaskan Arctic Tundra, found

no lasting effects of C02 enrichment on the vegetation (Tissue & Oechel 1987, Oechel er

al. 1997a). The present consensus is that elevated carbon dioxide levels will have the
greatest potential effects on vegetation in water stressed habitats (Chaves & Pereira
1992), and crop (Dahlman 1993) and timber (Eamus & Jarvis 1989) production. Other
studies have shown variable long-term effects at the community level (Bazzaz 1990,
Culotta 1995, Koch & Mooney 1996). Overall, the conventional wisdom is that CO2
enrichment alone will have little to no effect in the Arctic (Billings 1995, Oechel et al.

1997a).

1.2.2-2 Climate and Plants

Climate has been attributed to be the ultimate and proximate cause of vegetation
communities and in turn the animal communities that inhabit them. Many classiﬁcation
schemes use climate as the key to determining vegetation; the most applied of these are
the Holdridge (1947) life zone system and the Box (1981) model. These schemes predict
vegetation on the basis of variants of precipitation and temperature. When considering
the effects of climate on the biota, the range of variability is also of great importance
because it is often extreme events that alter a species distribution and ultimately
communities (Grime 1990, Sparks & Carey 1995).

Temperature is important for chemical processes and nearly all plants are
exotherrns, therefore, their metabolic rates are closely dependent on the temperature of
their environment. Plants have evolved various physiological adaptations to maintain
similar metabolic rates across the latitudinal zones. On a daily basis plants modify
internal temperature to various degrees by opening the stomata and altering the angle that

they intercept solar radiation. Long-term morphological adaptations allow plants to

change the energy balance between themselves and the environment in order to gain heat
or dissipate heat (Bliss 1962). Plants have also evolved similar reaction rates by varying
enzyme types and concentrations as temperatures change in order to maintain more
constant reaction rates (Christophersen & Larcher 1973, Chapin & Shaver 1985a). These
adaptations become more extensive and elaborate in more extreme environments such as
the Arctic (Billings 1974, also see section 1.2.3-l). In fact, the vegetation of the tundra
can be as productive on a daily basis as the vegetation of temperate regions because of
adaptations that allow plants to function similarly across a wide range of temperatures
(Webber 1978, Bliss 1988). Each species has its own unique suite of adaptations to cope
with temperature; therefore, each species has its own ecological temperature optimum
and range (Larcher 1975). However, there are limits that higher plants have yet to
overcome. These are the cardinal points 0°C and 35°C. Furthermore, the proximate
short-term response and the ultimate long-term response of a species to new temperature
regimes are not necessarily similar due to their prior adaptive strategies (Chapin 1987,
Ozenda & Borel 1989). The combination of these factors and others creates a situation
where each species has the potential to respond uniquely to changes in environmental
conditions and prediction of a species’ new realized niche is a daunting task (KOrner
1994, Billings 1997, Huntley & Cramer 1997).

Many researchers predict major changes in species diversity as a result of climate
change (Peters & Lovejoy 1992). Fundamentally a population has three potential
responses when presented with a change in the environment beyond its tolerance range or
ability to acclimate: adapt, migrate, or go extinct (Stonehouse 1989, Holt 1990).

Presumably species respond in the above order. Species have the potential to cope with

ncw simian
that: It
plasticity
west!
can cvolt
classic I
adtance
able to
& Kinl
habitat

the fin

new situations and it is likely that in the short-term many species will manage with new
climate regimes as they do with interannual variability, although this is dependent on the
plasticity of the species. In the long-term species will presumably be out competed by
species previously adapted to new climatic conditions (from lower latitudes) before they
can evolve new adaptations. In this situation the potential response is to migrate. The
classic example of migrations of this type is the progression of species during glacial
advance and retreat. However, many species possess low dispersal rates and may not be
able to change ranges at the same velocity as the change in climate (Davis 1989, Solomon
& Kirilenko 1997). There is also a possibility that analogous high alpine and arctic
habitats may no longer exist leaving a species with no place to migrate to. This leaves
the ﬁnal potential response, species that cannot cope or adapt to new conditions or
migrate to favorable habitats will become extinct.

The proximate effect of increased temperature will presumably be changes in
species ﬁtness. It is believed that temperature is important enough that over the long-
terrn this will lead to an ultimate effect of changes in species distribution. Many cite
examples of migration during the ice ages as evidence that species respond to climate,
and it was believed that entire vegetation zones moved along with glacial retreat (Oosting
1956). This follows a Clementian view of a climax community where the climate
ultimately dictates the community in an orderly way (Clements 1916). Recent more
detailed studies Show that communities did not move en bloc; rather, species responded
individually, creating new communities as their individual distributions changed in a
Gleasonian way (Gleason 1926, Delcourt & Delcourt 1981, Davis 1989, Bartlein et al.

1997). There are many non-climatic factors that species respond to which can modify a

9

species response to climate; these include nutrient availability, succession, competition,
herbivory, and disease to name only a few. Also, species have a migration rate, climatic
sensitivity and internal resistance which are predetermined and often conﬁned by prior
adaptations and evolutionary history (vae & Hive 1974, Huntley 1991, Billings 1992,
Hoffmann & Parsons 1997). These factors in combination cause species to respond
uniquely to climate change. The Emanuel er al. model (1985), presented in section 1.1, is
illustrative of the potential for vegetation change but is fundamentally ﬂawed and limited

in its true predictive value because it does not account for the individualistic nature of

species responses.

1.2.3 The Arctic System

The nature of the arctic environment creates a Situation in which even modest
warming has the potential to greatly increase the habitability of the arctic climate because
temperatures are habitually close to the biological threshold of 0°C. Figure 1-2
graphically depicts the climate of Barrow. From an examination of the ﬁgure the
potential for the Arctic System to respond to warming can be identiﬁed. With modest
warnring the average daily range of temperatures may no longer overlap zero, snow melt
could occur earlier and begin to reaccumulate later, the winter’s frozen soil could melt
earlier, and the active layer may become thicker. Warming of this layer would increase
biological activity. The combination of these factors could create a longer and warmer
growing season and increase the availability of nutrients (Kane et al. 1991, Anderson
1991, Hobbie 1996, Anisimov et al. 1997). Each of these factors alone could effect the

system, and there is also the potential for synergism between these factors (Chapin 1984,

10

Parsons et al. 1994). Additionally, the Arctic System is predicted to warm proportionally
more than other biomes for reasons presented in section 1.2.1-1. Thus, the biota of the
Arctic is predicted to respond ﬁrst and most to climatic warming (W ebber & Walker
1991). The combination of these factors makes the Arctic System an important and

useful location to study biotic response to climatic change.

  
 
    
   
 

 

2Ox102 F Redtetlon absorbed
g. nu! “mu” H limit; '"
.L 2: ......tnnIIIIIIIIHIIHIIIIllllllllllllllllllllllllllllllllm , I” ll '”"

 

 

Monthly eIr temperature range

J F M A M J J A 8 O N D
t J 1 1 1 l r r r r 1
Month

Figure 1-2. Diagram of mean, maximum and minimum temperatures, snow depth, active
layer thickness, and solar radiation at Barrow (from Chapin and Shaver 1985a).

1.2.3-1 Arctic Plant Adaptations
The arctic environment limits plant growth via low temperature, strong winds,
low light intensity, low nutrient availability, seasonal water stress, and short growing
seasons (Savile 1972, Bliss et al. 1973). These constraints reduce carbon gain, thus
hindering reproduction and productivity. This is one rationale, among many proposed,

for the simple community structure of the tundra system (Warren Wilson 1957, Walker

11

1995). An equally plausible reason for simple community structure is the relative youth
of the tundra biome, which also explains why the ﬂora is of diverse origins and has no
endemic genera and relatively few strictly arctic species compared to other biomes
(Dunbar 1968).

Tundra plant species often have wide geographic distributions, wide tolerances,
and many ecotypes (Billings 1997). Tundra plants are generally long-lived perennials,
use asexual and vegetative propagation, allocate large quantities of carbon to
reproduction, have large long-lived seed banks, and are polyploid (Johnson 1969, Savile
1972, Molau 1993a). These attributes are considered to be useful adaptations to life in
the tundra environment. The life form spectrum for the tundra is termed chamaephytic.
The spectrum is generally composed of 0% phanerophytes, 23% chameophytes, 61%
hemicryptophytes, 15% cryptophytes, and less than 1% therophytes (Oosting 1956).
arctic species often fall into four groups based on their distribution: hyperarctic (inhabit
the high arctic), eurarctic (inhabit the entire arctic), hemiarctic (inhabit the mid arctic, not
the extremes of high or low), and hyparctic (inhabit the low arctic and taiga) (Chemov
1985). These groupings overlap considerably with the groupings of characteristic species
within Young’s zones, which are based on climate and species composition (Young
1971). High arctic communities and species have been shown to respond more to
temperature manipulation than low arctic communities and species (W ookey et al. 1993).

Many arctic plants have evolved ways to absorb heat and maintain considerably
higher tissue temperatures than their local environment. Some of these morphological
adaptations are, short stature, maintaining dead parts to reduce wind, and dark

pigmentation (Savile 1972, M¢lgaard 1982). In order to compensate for low rates of

12

metabolic and physiological process at low temperatures Arctic plants have evolved high
enzyme concennations which enable them to maximize metabolic and physiological
processes including photosynthesis, nutrient absorption, and growth at temperatures far
lower than related temperate species (Larcher 1975, Heide 1983, Chapin & Shaver
1985b). Due to morphological and physiological adaptations, arctic species are generally
believed to be more limited by indirect effects of temperature on other abiotic factors
namely nutrient availability and length of growing season than direct temperature effects
on plant physiology (Chapin 1984, Chapin & Shaver 1985a).

The low level of plant nutrients in arctic environments is generally believed to be
due to slow decomposition and tumover (Swift et al. 1979, Hobbie 1996). Low soil
fertility is veriﬁed by many fertilization experiments and is supported by the dominance
of species with high root to shoot ratios (Babb & Whitfield 1977, Shaver & Chapin 1986,
Henry et al. 1986, Jonasson 1992). The problem of low nutrients is compounded by the
fact that tundra plants have higher than average nutrient demands due to their high
enzyme and lipid concentrations (Chapin & Shaver 1985a). In situations where nutrients
and water are not limiting, the largest constraint on productivity is the length of the
growing season. On a daily basis, the relative growth rate (RGR) of Eriophorum
angustifolium of 128 mg g’1 d'1 is greater than the range for temperate plants where RGR
levels range from 16 to 60 mg g‘1 d'l (Chapin & Shaver 1985a). Tundra plants are often
evergreen or semi-evergreen (often referred to as Wintergreen), and often preform
vegetative and ﬂowering buds up to several years in advance (Sarensen 1941). This may
be an adaptation to the short growing season. A further adaptation my be the general lack

protective scales or hard parts over buds so that they can readily expand at the onset of

13

snowmelt and begin to grow while their roots are still frozen (Savile 1972, Shaver &
Kummerow 1992).

Two often distinctly different growth strategies have been recognized in the
Arctic. These are known as periodic and aperiodic growth (Serensen 1941). Species that
show periodic growth are considered to be less receptive to changes in heat accumulation
and to generally grow to a predetermined size regardless of a current season’s climate.
Species that show aperiodic growth commonly respond directly to climate and may thus
be thought to take advantage of warmer and especially late season temperatures.
Aperiodic growth may allow a species to fully utilize the growth potential of a season;
however, the individual could be more susceptible to harsh summer conditions and winter
injury (Sorensen 1941, Savile 1972). The periodic growth strategy may be considered to
be more conservative and to reduce the risk of damage due to harsh weather. Periodic

growth is more common in the Arctic (Savile 1972).

1.2.3-2 Barrow Alaska

Barrow was chosen as the ﬁeld site location because it has cold summers, a long
climatic record, site management and protection, and an ecological database (Shaver
1996). The vegetation of the area has been well described by many including Britton
(1957), Cantlon (1961), and Webber (1978). It was the site for the Tundra Biome
program of the International Biological Programme (Brown et al. 1980). Barrow has
been the home of the Naval Arctic Research Laboratory (NARL) and its succeeding
institutions since 1947 and has one of the longest and richest histories of research of any

location in the Arctic making it one of the best known ecosystems of any in the world

14

(Reed & Ronhovde 1971). A Climate Diagnostic Laboratory run by NOAA is located
within 3 km of the site and provides historical climate data as well as detailed current
conditions and is the site of one of the longest C02 and greenhouse gas records in the
world (Hofmann et al. 1996). Recently the community of Barrow established the Barrow
Environmental Observatory (BBC) to preserve a large tract of tundra for future research.
Thus, because of its huge potential to respond to modest warming, its inherent simplicity,
and its historical record, Barrow serves as an ideal location to study species response to

climatic change.

1.3 RESEARCH FRAMEWORK

The results presented in this thesis are a subset of two distinct yet interacting
research programs, Arctic Systems Science (ARCSS) and the International Tundra
Experiment (ITEX). All work for the thesis was completed as a part of the Arctic
Ecology Lab (AEL) on Michigan State University. The logos of these 3 research groups

are presented in Figure 1-3.

1.3.1 Arctic System Science

The Arctic System Science (ARCSS) Program is a subprogram of Polar Programs
within the National Science Foundation (ARCUS 1993). ARCSS takes a whole-system
approach to understanding the response of the Arctic System to global change and is
particularly concerned with the mechanisms and consequences of the ampliﬁed response

of the high latitudes to greenhouse warming. A principal goal of ARCSS is to enable the

15

prediction of the future state of the Arctic System, on seasonal to century time scales, by
integrating observations, process research, modeling and assessment. This ITEX project
is most closely connected to the Land / Atmosphere / Ice Interactions (LAII) component
of ARCSS. The response of tundra vegetation to warming documented in this thesis will
ultimately be incorporated into models that attempt to predict vegetation response to

global change.

1.3.2 The Arctic Ecology Lab

The Arctic Ecology Lab (AEL) is housed in North Kedzie Hall on the Michigan
State University (MSU) Campus. MSU is a large, well-equipped research university with
superior library, computational, and soil / stable isotope analyses facilities. The
knowledge resource at MSU is extensive and includes an interactive network of 83
Ecology and Evolutionary Biology faculty members representing 12 college departments
from Anthropology to Zoology. The AEL occupies ﬁve ofﬁces and houses 1 faculty
member, 2-3 graduate students, and 3-7 undergraduate assistants. Each Ofﬁce has its own
phone, computer, and printer. The computer facilities were designed for graphics and
statistical analyses and include a scanner and color laser printer. All equipment is
connected to the web and locally networked. The AEL is an active collaborator with the
Computational Ecology and Bioinformatics Laboratory (CEBL). CEBL is designed to
provide computational facility to conduct spatial analysis research and use intensive
mathematical models to address global and complex systems analysis. The AEL library
contains over 15,000 volumes and 30,000 reprint with emphases in Cold Regions

Ecology, Global Change, and Botany; it also houses an extensive map and aerial

l6

photograph collection of the Northern Alaska, particularly the Barrow area and the

National Petroleum Reserve.

1.3.3 The International Tundra Experiment

The International Tundra Experiment (ITEX) is a collaborative effort involving
scientists from 11 countries including all the Arctic Nations (Figure 1-4). ITEX seeks to
examine the response of circumpolar cold adapted plant species to environmental change,
speciﬁcally to an increase in summer temperature. Empirical knowledge based on
experiments coupled with available evolutionary history, ecology, and genetics was
chosen as the best way to predict species response to climate change. The ITEX research
model combines long-term and short-term experimentation with monitoring and has the
elegance and simplicity called for to understand ecosystem response and vulnerability to
change (Tilrnan 1987, Rastetter 1996). The experiment is designed to examine the
effects of temperature change; maximize geographic representation, by minimizing
technical and equipment requirements; be long-term; focus primarily on species; and, if
resources permit, allow for genetic and system level studies (Molau & Molgaard 1996).
Participation may be at several levels of complexity and sophistication depending on
interests and available funding support. Each ITEX site operates the base experiment,
which uses small open-top green houses to warm the tundra. These passive chambers
effect plant growth and phenology in a variety of ways (Marion et al. 1997, Henry &
Molau 1997). Collectively the ITEX network is able to pool its data sets to examine
vegetation response at varying levels, for example genetics (from ecotype to functional

type), across space (from habitats to ecosystems) and over time (Walker 1996a).

17

 

Figure I-3. Logos of the research efforts with which this project is networked: (A) the
Arctic Ecology Lab, (B) Arctic System Science and (B) the International Tundra
Experiment.

1.3.3-1 The Network and its History
In 1990 global warming research focused on responses and roles of the abiotic

factors of ice, atmosphere and ocean in the Arctic System or on broad ecosystem
functions. By doing so, the research omitted organism speciﬁc responses to predicted
warming (W ebber 1990). Without this later information it is impossible to predict the
composition of future biotic communities and living resources. Therefore, a workshop
was held to explore the development of an international, Arctic-wide network of
observations and experiments to look speciﬁcally at the response of tundra plants to
climate warming. The workShOp involved 50 scientists from many nations and resulted
in a project and an experimental protocol (Webber & Walker 1991, Molau 1993b).
The ﬁrst ofﬁcial ITEX measurements were made in 1992 and now the network comprises
26 Sites. ITEX is coordinated from the Danish Polar Center, Copenhagen. U.S. ITEX
projects are part of the NSF Arctic System Science (ARCSS) Program of the U.S. Global
Change Research Program. ITEX scientists meet annually to share experiences and
coordinate efforts. They met in 1996 at the National Center for Ecological Synthesis,
Santa Barbara (NCEAS), CA where they used meta-analysis in an innovative way to
integrate pooled data sets (Gurevitch & Hedges 1993). ITEX will continue this approach
to synthesis as the data set grows. ITEX scientists are publishing in the open literature,
including a special issue of Global Change Biology (Henry & Molau 1997). The ITEX
modus operandi for studying biotic response to climatic variation is well regarded and is
used as a model for other studies. For example, the Circumpolar Active Layer
Monitoring program (CALM) has modeled their coordinated effort after ITEX (Brown

1997).

19

 

Figure 14. Map of the ITEX ﬁeld sites. Site 13 is Barrow (Molau & Molgaard 1996).

20

1.3.3-2 The Chamber

A passive open-top chamber was chosen as the preferred method to warm the
tundra because of its low cost, its ease of replication, and freedom from the need of
mechanical investment or power supply (Figure 1-5). An open-top was chosen to lower
temperature extremes and allow more direct solar radiation, more natural levels of
humidity and gases such as C02, direct entry of precipitation, and easier access of
pollinators and herbivores. The structural soundness of the chambers enables them to
hold up to the rigorous of the arctic climate and be used for many years. Due to justiﬁed
concerns about the use of chambers (Kennedy 1995), an extensive documentation of

chamber performance was done and is presented in Chapter III.

 

Figure 1-5. An open-top chamber (OTC).

21

1.4 QUESTIONS AND HYPOTHESES

For clarity, the responses of plants to temperature are discussed under three
themes: phenology, plant vigor, and species patterns. Each of these themes are discussed

separately in terms of underlying questions, speciﬁc hypotheses, and the rationale.

1.4.1 Phenology Response
mm In what way and to what extent will the phenology of selected plants respond

to variations of growing season temperature and increased canopy temperature?

M13; Flowering will occur earlier under warmer conditions.

MES—11 Flowering will occur at approximately the same degree day threshold.

Rationale: Species in sheltered depressions and on the south facing slopes ﬂower earlier
than individuals in the adjacent environment due to warmer microclimates (Sorensen
1941, Bliss 1962, Savile 1972). Observations along climatic gradients also Show that a
species ﬂowers earlier at the warm ends of its climatic range (Serensen 1941 , Walker &
Webber 1979, Shaver et al. 1986). Although arctic plants have adapted to low
temperature, their tissue temperatures are generally below their optima for most
physiological processes (Chapin & Shaver 1985a). Temperature is believed to be critical
to growth and development and degree day accumulations have been shown to be
strongly correlated with onset of phenological events (Lieth 1974, Wielgolaski &
Karenlarnpi 1975, Shaver & Kummerow 1992). Therefore we would expect Species to

respond to warmer microclimates and warmer ﬁeld seasons with earlier onset of

22

phenologic events, namely ﬂowering (Dennis 1969, Walker et al. 1995, Molgaard &

Christensen 1997, Walker 1997).

1.4.2 Vigor Response
Mn; By how much and in what way will the vigor of selected plants respond to

variations of growing season temperature and increased canopy temperature?

W Plants will produce more ﬂowers under warmer conditions.
W The height of inﬂorescence will be taller under warmer conditions.

W The length of leaves will be longer under warmer conditions.

Rationale: Although arctic plants are generally believed to be acclimated to lower
temperatures, their optima are above most prevailing tissue temperatures. Therefore, we
would expect greater carbon accumulation and overall vigor of a plant with increased
temperatures (Chapin & Shaver 1985a). Sexual reproduction is only occasional
successful in the arctic environment, but is believed to be important. If a plant has
additional photosynthate, it may be allocated to reproduction and consequently ﬂower
more (Bliss 1988, Molau 1993a, Jonsdottir 1995). Arctic plants are generally short in
stature and live near the ground’s surface where one might reason that biotemperatures
can be maximized (Warren Wilson 1957, Hansen 1973, Bliss 1988). Nevertheless, there
is an advantage to increasing the length of the inﬂorescence above the plant canopy to
enhance dispersal potential (Savile 1972). With warmer air temperatures plants are
predicted to increase in stature. The leaf length is also predicted to increase because of
increased overall health of the plant, and the potential for greater light competition

(Walker 1986, Callaghan et al. 1989, Chapin et al. 1995, Harte & Shaw 1995).

23

Additionally, specimens are generally taller and have longer leaves in their warmer

ranges or microclimates (Sorensen 1941, Dennis 1968, KOrner & Larcher 1988).

1.4.3 Species Patterns of Response
Mien; Are there species responses to temperature that can be generalized or do all

species respond uniquely?

Hypothﬁis 3g: Short-term plant phenologic response to canopy warming will mirror
responses of the controls observed in warm years.

W Short-term response of plant vigor observed in the chambers will mirror
responses of controls observed in warm years.

Hmthesis 4: Plant species will respond individualistically to temperature.

Ratignale: Every species has evolved its own way to cope with the climate and each has
approached the problem from very diverse historical origins (Savile 1972, Billings 1992).
Studies have shown that arctic species respond to external stimuli individually (Chapin &
Shaver 1985b, Oberbauer et al. 1986). Therefore, it is predicted that species will also
respond uniquely to warming.

The results of hypotheses 3a and 3b will determine the efﬁcacy of the chambers
as a temperature enhancement device in terms of plant response. If the chambers main
effect is to raise temperature, then the species’ short-term response to chambers should be
similar to a species’ interannual variability in ﬂowering. If changes in vigor are strongly
related to temperature and the chambers do not produce undesirable secondary effects,
then Short-term response to warming due to chambers should mirror the response due to

interannual variability in seasonal temperature.

24

Chapter II
METHODS

11.1 STUDY AREA

The research area is located within the Barrow Environmental Observatory (BEO)
approximately 8 km north the village of Barrow, Alaska on the Banow peninsula
(71°18’N, 156°40’W)(Figure II-l). The BEO is a protected tundra landscape covering 15
square kilometers owned by the Ukpeagvik Inupiat Corporation (UIC) which is leased to
and managed by the Barrow Arctic Science Consortium (BASC). The Barrow Region
has a long and rich research history (see section 1.2.3-2). The ﬁeld site is located within
an Arctic System Science (ARCSS) grid and within 100m of an ITEX site in a dry heath
community (Figure 112, Bay 1995,1996, Walker 1997). There are several ARCSS grids
throughout the Arctic that monitor long-term change in variables such as snow depth,
active layer progression, and plant community composition (Brown 1997).

The Barrow Peninsula is a spit of land surrounded by the Chukchi sea on the west,
Elson Lagoon on the east, and the Arctic Ocean to the north. The Peninsula is roughly
25,000 years old (Brown & Sellrnann 1973). The coastal tundra in this region is
dominated by a pattern of ice wedge polygons, and shallow lakes. The research area is
contained within a wet meadow community situated on the north eastern end of Central
Marsh in a transition zone between the marsh proper and a former raised beach ridge
surrounding the marsh on the north side (Figure 11-2). The elevation of the site above
mean sea level is 3 i 0.5 m and has a ﬁne silt substrate and histic pergelic cryaquept soils
with very poor drainage (J. Bockheim, personal communication). The ﬁeld site is on a

recovered former vehicle track that ran along the edge of Central Marsh.

25

 

I l

f
AaCTIC OCEA
a. (4" ., .

        

 

 

 

 

 

“‘5'" Polygonized Moist Meadow ‘5'“

Figure IL]. Map of Alaska and the Barrow Peninsula showing the ITEX ﬁeld sites and
other historical research locations.

 

 

 

 

 

 

 

 

 

Figure 11-2. Map of the research area, showing the ARCSS lxlkm grid (crosses), the
ﬁeld sites, beach ridge, and other prominent features. The Boundary of the Barrow
Environmental Observatory (BEO) is shown on inset (I-Iinkel et al. 1996).

26

The climate of Barrow consists of long cold winters and short cool summers
during which the temperature can fall below zero on any day (Table 11-1, Figure 1-3).
The sun does not rise from November 18 to January 24 and rise is 24 hours a day from
May 10 to August 2. The snow free period is variable but generally begins in early June
and continues until early September during which time an average of 251 degree days are
accrued (Brown et al.1980). The summers are generally cloudy, foggy and wet with a
summer average of over 80% humidity during which 37% of the annual precipitation is

received (Brown et al. 1980).

Table 11-1. Climate data for the Barrow region (Brown et al. 1980).

 

 

Temp Precip Windspeed Solar Day

(‘0 (mm) (m s") radiation Length
(M1 In" day") (M
Jan -25.9 5.8 5.0 0.0 0.7
-28.1 5.1 4.9 1.6 6.8
-26.2 4.8 5.0 7.4 l 1.7
Apr -l8.3 5.3 5.2 15.5 16.7
May -72 43 52 21.9 23.1
June 0.6 8.9 5.1 23.0 24.0
July 3.7 22.4 52 18.5 24.0
3.1 26.4 5.5 10.8 19.0
Sept as 14.7 5.9 5.0 13.4
Oct -9.3 14.0 6.0 l .7 8.6
Nov - l 8.1 7.6 5.6 0.2 2.4
Do: ~24.6 4.8 5.0 0.0 0.0

Year 12.6 124.1 5 3

 

The vegetation of the region is coastal tundra dominated by grarninoids, most
notably Carex starts and Eriophorum species. The Barrow tundra is acidic and contains a
major bryophyte component and an abundance of lichens. Several relatively distinct
vegetation associations are present in the region including: Luzula heath, Salix heath,
Carex-Poa meadow, Carex-Oncophorus meadow, Dupontia meadow, Carex-Eriophorum
meadow, Arctophila pond margin, and Cochlearia meadow (Brown et al. 1980). The

location of the ﬁeld site is within a Carex-Eriophorum meadow. The species

27

composition determined by the point frame method (Walker 1996b) is given in Table II-2

for the combined experimental open-top chamber plots and control plots in the site.

Table II-2. Species composition of the ﬁeld plots.

 

Percent Percent
VASCULAR PLANTS Covet VASCULAR PLANTS Cover
Gum m 40.50 Caduceus Mulls 0.41
Ericphoum m 13.51 Salbr Malia 0.36
Dtptanla W 10.97 Draba hem 0.37
Stalled: beta 5.75 Parasites In’gidus 0.37
Sam Merlin 5.55 mm nivalis 0.09
Edophorun m 3.47 W pypamws 0.05
Carnation) 69W» 3.06 Alopecums alpinw 0.04
Suing: com 3.06 Draba mm 0.04
Gardenias preterm 2.55 W m 0.03
Camswepam 1.56 Eriaphonm W 0.03
Handiloe pawﬂom 1.49 Salaam: 0.01
Saxllmga folioloea 1.30 Suing. conspire“ 0.00
Suffrage W 1.24
Poe am 1.22 LITTER 25.50
Jutcus biglumis 0.74 BRYOPHYTE 43.20
Coleman‘s holmﬁ 0.64 UCHEN 2.20
Luzula arctic: 0.61 ALGA a MUSHROOM 0.30
Luzula contra 0.46 BARE GROUND 0.10
Stellar-in mm 0.46

 

11.1.1 Site Establishment

The data presented in this thesis were recorded during the ﬁeld seasons of 1995
and 1996. The ﬁeld site was established on June 25, 1995 along a snow bed on the
northeastern side of Central Marsh. Areas of high plant species diversity containing
preferred species were chosen for plot establishment. Twenty-four controls and 24
experimental designations were determined randomly from the predetermined plots. The
plots were located 16 m from the retreating snow bed and formed a near linear pattern
due to the nature of the snow bed (Figure II-3). The plots were chosen as close to one
another as was possible to minimize variation in elevation, hydrology, soils, and plant
community composition. The site spans a length of less than 300 meters G-‘igure II-4).

Chambers were installed on the day of site establishment. Control plots were marked by

28

using wooden stakes and string to deliminate a lxlm square of tundra. A wooden stake
was installed near each chamber for identiﬁcation purposes. Metal tags were used to
mark each plot in addition to using waterproof markers. A path was established along the
south edge of the plots to minimize disturbance due to foot trafﬁc.

Chambers were removed on August 22, 1995 to avoid damage from local
snowmobiles and to avoid unnatural build up of snow in the chambers. They were
reinstalled between June 4 and 14, 1996 (the day after an individual plot became

completely snow free) and removed on August 21, 1996.

 

 

Figure II-3. The ﬁeld site shortly after establishment in 1995 (courtesy of Christian Bay).

29

I ma
£13.32.”

 

 

 

Figure 114. Map of ﬁeld site showing each plot.

30

11.2 Antone DOCUMENTATION & CHAMBER PERFORMANCE

11.2.1 Chamber Design

The open-top chambers (OTC ’s) are hexagonal with sloping sides constructed of
Sun-Lite HPTM ﬁberglass (Solar Components Corporation, Manchester, NH). This
material is commonly used in many horticultural settings due to its high solar
transmittance in the visible wavelengths (86%) and low transmittance in the infra-red
range (<5%) (Molau & Molgaard 1996). The height of the chamber is 35cm and the
distance between the parallel sides is 103cm at the base and 60cm at the top (Figure 11-5).
The open-top reduces shading effects and allows ventilation and access by pollinators.
Marion er al. (1993 & 1997) documented the general performance of the ITEX OTCS and

a detailed documentation of the chamber performance at Barrow is given in Chapter 111.

11.2.2 Climate Monitoring
11.2.2-l Macroclimate Monitoring
The macroclimate data for Barrow, Alaska were obtained from the National
Ocean and Atmospheric Association’s (N 0AA) nearby climate station. The station is
staffed year round and collects general climate data as well as several trace gas

concentrations (Stone et al. 1996).

11.2.2-2 Microclimate Monitoring
Both Hobo® and StowAwayTM data loggers manufactured by Onset Inc.
(Appendix Figures A-l - A-5) collected temperature and relative humidity data within

chambers and over control plots during the snow free season only. The StowAwayTM

31

70m

Side 35
View a"

 

Fl

 

 

 

 

 

 

 

 

 

 

 

 

 

It 120 an ,I l
A
31:51! 60 cm 103 cm
V
T
Individual
Fiberglass 40.50".
Panel 7y
.1.
I< m .m >1

Figure 11-5. The design of the open-top chambers.

32

data loggers are an expanded version of a Hobo® data logger that includes more memory .
and a more frequent recording interval. Data loggers were set to record every 12 to 80
minutes depending on the sensor type from the date of plot establishment until August
18. The placement of temperature data loggers was determined randomly each year;
relative humidity logger placement was determined randomly from the plots already
containing temperature loggers. Loggers or sensors were housed in “gill six plate”
thermistor shields at approximately 13cm above the ground in the most northerly comer

of the chamber (see Figure II-6A).

11.2.2-2.1 Horizontal and Vertical Temperature Distribution

A secondary monitoring program was established near the ﬁeld site to collect
more detailed data on the horizontal and vertical disnibution of warming within the
chambers. The horizontal distribution of warming within the chamber was established by
placing two arrays of thermistors at 1cm above soil surface within two chambers and four
thermistors nearby outside the chambers as the controls. The spacing for both arrays was
a thermistor at: approximately 10cm from the north edge of the chamber, half way
between the north edge thermistor and the center, the center, half way between the center
and the south edge thermistor, and approximately 10cm from the south edge (Figure 11-
6). For the vertical distribution of warming there were no replicates due to a limited
availability of sensors. In the one chamber and the nearby control, the spacing of the
sensors was: 1, 6, 11, 16, 21, 26, 31, and 36cm above the soil surface (Figure II-6). All

sensors were operated at a recording interval of 9 minutes from J une 20 to July 28, 1996

33

1 r”,

_- Gill Six Plate
SensorArray \

»

;'\'YZNY: 7'1

h"..- 3298

Figure II-6. The spatial arrangement of temperature probes: (A) horizontal distribution
and (B) vertical distribution.

 

34

for both the horizontal and vertical sampling. Sensors were shielded from direct sunlight

with shelters made from inverted polystyrene foam drinking cups (Figure II-6).

11.2.2-2.2 Light Distribution

The distribution of light entering the chamber was monitored with StowAwayTM
light sensors (Appendix Figure A-5). Four sensors were placed within one chamber.
Two sensors were placed within the center of the chamber and one approximately half
way between the center sensors and the chamber edge for both the north and south edges
(Figure II-7). Two sensors were placed outside the chambers as controls. All sensors

were run from August 1 to 21, 1996 at a recording interval of 5 minutes.

 

Figure II-7. The placement of four light sensors within a chamber.

35

11.2.3 Below Ground Monitoring
11.2.3-1 Soil Temperature
Soil temperature was monitored with the use of both Hobo® and StowAwayTM
data loggers with sensors placed in the ground at a spacing of 1, 5, 10, and 15cm beneath
the surface (Figure II-8). The loggers were in operation from July 10 to August 18, 1996
at a 16 minute recording interval. The same sampling protocol was used to monitor the
soil temperature from an ITEX ﬁeld site in a dry heath community within 100 m of the

wet meadow ﬁeld site (Walker 1997).

of
I

t
3171' 2“. _. .
. «» .

* Gi'I'I‘Six Plate '

 

Figure II-8. The chamber within which soil temperature monitoring was conducted.

36

H.3.3-2 Active Layer Monitoring

The active layer thickness was measured to the nearest cm by thrusting a small
metal rod into the ground surface until frozen ground was reached (Figure II-9). The
active layer thickness was recorded daily at the beginning of the thaw season. For the
remainder of the ﬁeld season measurements were made weekly in 1995 and every 10
days in 1996. The active layer thickness for each plot was determined by probing outside
the control plot within 30cm of the four corners or by probing the center of the plot in
OTCS. Previous holes were not used for later probings because of the potential for

unrepresentative depths of thaw caused by heat sinks from water percolation and air

movement within the former holes (Hinkel er al. 1997).

    

Figure II-9. Probing the active layer.

37

11.2.4 Degree Day Calculation

In this thesis, degree days and thawing degree days are used synonymously. The
traditional growing degree days based on 5°C was not used because in Barrow 0°C is
generally the cardinal temperature at which growth begins (Dennis et al. 1978). In
temperate or tropical sub tropical regions the cardinal temperatures are generally 5°C or
10°C respectively. In order to avoid confusion the term thawing degree days is used
rather than growing degree days.

Thawing degree day accumulation from snow melt ('I'DD“) was calculated by
using the temperature data collected (between 12 and 80 minute intervals depending on
instrumentation). Plots were established after snow melt in 1995, therefore degree day
accumulations from snow melt until plot establishment were estimated by calculating the
thawing degree day accumulation from standard NOAA screen temperature data. This
estimate was necessary because the site was established after the estimated snow melt
date of June 19'”. The estimated TDD* is likely within ﬁve degree days of the true
number. August 18 was chosen as an ending date for meaningful degree day
accumulation. This is based on the observation that nearly all Barrow species begin
dormancy or senescence in mid August (personal observation, Miller et al. 1980). Also,
August 18 was within a week of when the end of season plant measures, described in the

next section, were collected both years.

38

11.3 VASCULAR PLANT MONrromNG

Individuals of each species present were permanently tagged in each plot. In
1995 galvanized nails were used to mark the individuals; the nails were later replaced
with wooden sticks in 1996 because they were deemed easier to use and less likely to
alter the canopy thermal regime. The ﬁrst three individuals of a species to turn green
within a plot were tagged in 1995 and have been continually monitored ever since (Figure
II-10). During the ﬁeld season of 1996 a map of each individual’s location within each
plot was created. When individuals died or lost their tags, replacement individuals were
chosen. Preference was given to replacement individuals of good health and in an easy to
monitor location. When available three individuals of each species within a plot were

monitored at all times.

11.3.1 Phenologic Monitoring

Plant development was followed throughout the entire summer. Plant
phenophases were determined based on species morphology and ease of measurements
(Table 11-3) (Molau 1993b). Individuals were monitored daily at the beginning of the

season and every second or third day during times of slow change.

11.3.2 Vigor Monitoring

Vigor measurements such as height, length of longest leaf, and number of leaves
were measured only once at the end of each ﬁeld season. All measurements were taken
within approximately one week of August 18th on both years. For a complete listing of

all vigor measurements collected see Table Il-4 (Molau 1993b).

39

 

1, ' ~ ’.

Figure II-10. Representation of the plant monitoring scheme: (A) recording sheet, (B)
marked individual, and (C) map of plot.

 

 

Figure II-10. Continued:

41

 

Table II-3. Phenologic measures recorded for all species monitored.

 

 

 

3
3 .9
1 i i
r g .5 i E
i e E g g g g g E
s;;_ E . Eight
2 a a r a a e a i a
yweseeeseeeseees9695969596959695969596959695969598
£8 m . 1
Nopocumsalpim X X X
Wichita X X X X
Cdunagroeﬂsholnil X XX XX XX XX XX XX X
Gardnmlnepramls X XX XX XX XX XX
Carexstms X XX XX (XX, XX XX XX
Camswspaﬁncea X XX XX XX. XX XX XX
Cerasiutmbcerindanum X XX XX XX XX XX
CocNoaﬁaatﬁclMie X XX XX XX XX XX XX XX
Drabalactoa X XX XX XX XX XX XX
Drabamlcropeua X XX XX XX XX XX XX
Dtpmtiaﬁsheri X XX XX XX XX XX XX XX
Edophonunmsseolun X XX XX XX XX XX XX
Eﬁophmuntn'ste X XX XX XX XX XX XX
Edopharunschwchzed X XX XX XX XX XX XX
Hierochloe pawlflora X X X X X X X X X X X
Juncusbldunls X XX XX XX XX X XX
Luzrrlaarctica X XX XX XX XX XX XX
unmacmfma X XX XX XX XX XX XX
Petasltesfn‘gidus X XX XX XX XX XX XX
Poaarctica X XX XX XX XX XX XX XX
Rammulusnimlis X XX XX XX XX XX XX
Rarumuluspygnaus X XX XX XX XX XX XX
Salixmdchra X XX XX XX XX XX
Salixraandfolla X XX XX XX XX XX
Saxitragacaespltoea X X X X XX X X X
Saxifragacm X XX XX XX XX XX XX XX XX
Saxltragalollolosa X XX XX XX XX XX XX XX
Saxifragahieraclfolla X XX XX XX XX XX XX XX
Saxifragamrcdus X XX XX XX XX XX
Stellaﬁahwmm X X X
salarlalactea X XX XX XX XX XX XX

 

 

 

 

 

 

 

 

 

 

 

 

 

42

 

Table Il-4. Vigor measures recorded for all species monitored.

 

 

 

 

3 i
o 8 g
t 3 i i g g g E
yw95%9598%%9596%$959695%9596
M
Nopacmalplws X Xl
Wichita X X
Cdmoeﬂsholmﬂ X X XX
Cardaminepmtmsis XX XX XX XX XX
Camstans XX XX XX XX XX XX
Carexswspartuooa XX XX X
Wmoeerinpiarm XX XX
Cochloaria aﬂiclnalis X X X X X X
Drabalactea XX XX XX
Drabanicropeﬂa XX XX XX
Dtpmﬁaﬁshen' XX XX XX XX
Eriophorunmssodun XX XX XX
Eriophonantrisb XX XX XX XX XX
5:1meth XX ‘ xx xx
Hierochloe pawlﬂora X X X X X X
Jmewbiglunis xx x
Luzulaaretica XX XX X
museum“ XX XX X
Patasltostrigiws X XX XX
Poaarctica XX XX X
Ranmcuusnlmlis XX XX XX XX
Ranmculuspygneus XX XX XX
Salixprdchra XX XX
Salixratmdfolla XX XX
Saxilmgacaespltosa XX XX XX
Saxilragacamua XX XX XX XX XX XX
Saxlfragalollolosa XX XX XX XX XX
Saxitraga Hawaiians X X X X X X X X
Saxifraga hireulus X X X X X X
Sellariahumltusa XX XX
Stellarlalactea XX XX

 

 

 

 

 

 

 

 

 

43

 

11.4 DATA ANALYSIS

The basic experimental design was a non-orthogonal completely randomized
design with subsampling. The design was repeated in two ﬁeld seasons. For statistical
tests, plots were treated as the experimental units with multiple individuals within a plot
treated as observational units used to obtain an estimate of observational error. Proper
analyses of variance were run per species per variable. The homogeneity of the variances
for all analyses was tested. In many cases this assumption was violated and the analyses
were rerun with weighted averages adjusted by the population’s standard deviation. In no
cases were the conclusions different for the two analyses. Deviations from standard tests
are noted. Population averages and errors of the sample means were used for all graphs;
these calculations did not account for subsampling and weigh every individual
observation equally. An outcome was deemed statistically signiﬁcant if the probability
for a Type 1 Error was 5% or less. Calculation of descriptive statistics including means,
maxima, minima and standard deviations was done in Microsoft Excel 97 or SAS 6.12
(Microsoft 1997, SAS Institute 1990). All statistical tests were preformed in SAS by
using proc GLM or proc MIXED (SAS Institute 1990, 1996). Proc GLM uses a general
linear model and it was used to run analysis of variance with unbalanced data. Proc

MIXED was used to run analysis of variance on correlated data using repeated measures.

11.4.1 Active Layer Comparisons
The active layer thickness data were analyzed in proc MIXED using the
“repea ” statement (SAS Institute 1996). The active layer thickness at any given plot

was correlated with measurements at the same point taken at other times during that ﬁeld

season. Active layer thickness was not presumed correlated across years because of
differences in climatic attributes known to affect soil freezing such as winter air
temperatures and snow cover. The correlation pattern used was a continuous auto-
regressive model as a function of Julian days. The SAS code used was:

proc nixed data - aydata.act1ve;
class year treat julian plot;
nodel thaw = year treat plot year*treat
julian(year) treat*julian(year);
repeated / type-sp(exp)(t1ae) subject=year*treat*plot ;
run;

ILAJBIHantCXanaﬁhons

Analyses of the response variables were run identically for all species.
IExennﬂanyanahnmstwenznnrnaexanﬁnernaﬁsﬁcalsunuﬁmance1whenrﬁfﬁuencesin
variance were accounted for. These analyses were never signiﬁcantly different from the
original analyses and are therefore not presented.

The standard analysis run on each species was an analysis of variance run
accounting for uneven replication numbers and subsampling sizes. The SAS code for this
analysis was:

proc glm data = nydata.speciesdb;
class year plot treat;
nodel <RESPONSE VARIABLE> = year treat treat*year

plot(treat*year);
test h = year treat treat*year e = plot(treat*year);
by 090599;
means treat*year;
lsneans treat*year lstderr e=plot(treat*year);
run;

45

The standard analyses nm all species at once to determine overall signiﬁcance
was an analysis of variance run in proc MDIED. These analyses also accounted for

unbalanced sample sizes and subsampling. The SAS code for the overall tests was:

proc mixed data = Iydata.spec1esdb;
class year treat plot genspp;
model <RESPONSE VARIABLE> = year treat genspp
year*treat year*genspp treat*genspp
year*treat*genspp;
random plot(genspp) plot(year) plot(treat)
plot(year*treat) plot(year*genspp)
plot(treat*genspp);
run;

The analysis run to compare the 1995 open-top chamber (OTC) with the 1996
control was used analysis of variance with unbalanced sample sizes. The “year” and
“treatment” groups were combined to make four separate groups and each of these
populations was tested to determine if the sample means were equal. Multiple
comparisons were performed with Fisher’s LSD (least signiﬁcant difference) and
Tukey’s HSD (honestly signiﬁcant difference) (SAS Institute 1990). In all cases the
signiﬁcances were similar for the two analyses. The SAS code used to compare the 1995
OTC with the 1996 control was:

proc glm data = mydata.speciesdb;
class txyear plot;
model <RESPONSE VARIABLE> = txyear plot(txyear);
means txyear I e=plot(txyear) LSD lines;

means txyear I e=plot(txyear) tukey lines;

by 090599;
run;
(Note the variables treatment and year were combined in order to easily run this analysis)

The percentage of individuals ﬂowering (presented in section IV.2.1) was
calculated as the ratio of the number of individuals for which the ﬂower opened to the

number of individuals monitored. Individuals that were found late in the season because

of the presence of an inﬂorescence were not included in this percentage. An analysis of
variance was also run on this data set.

The percentage of individuals responding similarly (reported in section IV.4.3)
were calculated by adding up the category presented from Tables IV -2 and IV -3 and

dividing by the total.

47

Chapter!!!

ABIOTIC DOCUMENTATION AND
CHAMBER PERFORMANCE

In this chapter the results of a detailed documentation of the abiotic environment
surrounding the vegetation is presented. In addition the performance of the chambers is

described in order to access the efﬁcacy of the open-top chamber as a warming device.

111.1 CLIMAT'IC DATA

The 1996 ﬁeld season was warmer than that of 1995. The temperature 13cm
above the control plots was, on average, l.2°C warmer (Table III-1). The ﬁeld season of
1996 received almost twice as much precipitation as that of 1995 (Table III-1). It should
be noted that the increased precipitation in 1995 might not have resulted in an increase in
soil moisture because there may have been differences in runoff and evapotranspiration
between the two years. The daily average maximum and minimum temperatures are
given in Appendix Table A- l: the temperature from the nearby NOAA meteorological
screen at 2 m above ground are also presented for comparison. It was found that plant
canopy temperatures of the control plots during summer are generally 0.5-1.0 °C warmer
than screen temperature.

On average, over the entire monitoring period, the chambers warmed the plant
canopy 19°C in 1995 and 1.4°C in 1996 in the monitored chambers. The relative

humidity was 7.3 % less than that of the controls in year 1996: this is commensurate with

48

an expected drop in relative humidity with increase in temperature. A more detailed

description of the chamber performance is given in the next section.

Table III- 1. Summer average, maximum, and minimum daily temperature and relative
humidity of chambers and controls and climatological data of the nearby NOAA station
from snowmelt until August 18till for 1995 and 1996.

 

 

NOAA Control OTC

Precip Temperature Relative Temperature Relative Temperature Relative

(an) (C) W (C) W (C) Hmldty
year mmnﬁmwnﬂmmnﬂmwnﬂmmnﬂmmm
1995 18 3.2 0.0 1.4 89.4 998 -— 3.6 7.7 0.4 -- -- -- 5.5 122 0.9 -- -- --
1996 33 3.7 7.7 1.0 89.1 as; -- 4.8 9.4 0.7 89.1 95.8 00.5 6.2 may 76.4 69.2 03.1

-detenotavailable
111.2 CHAMBER PERFORMANCE

[11.2.1 Temperature and Humidity

The OTCS passively warm the plant canopy by acting as miniature greenhouses.
By their design, the OTCS are dependent on solar radiation for warming. On days of
dominant cloud cover the chambers only marginally warm the plant canopy, while on
sunny days the chambers substantially warm the canopy. During the summer “night,”
when the sun is at a low angle on the horizon, the chambers generally do not warm the
canopy, thus accentuating the normal diurnal pattern of warming that occurs in the
absence of a true night. This causes an increase in maximum daily temperatures but not
the daily minimums. As expected, the relative humidity follows changes in temperature
inversely. These patterns of warming and humidity are shown in Figure III-l for three
representative days. The daily course of warming in the chambers mirrors that of the

controls and is Strongly tied to incoming solar radiation. On all days the average

49

temperature within the chambers was greater than that of the controls. The spikes in the
temperature trace changes in cloud conditions. The chamber response is sensitive to
changes in radiation and shows a larger daily range of temperature than the control.
There is also a greater daily variation of temperature in the chambers than the controls

because the chambers warm substantially more on days with clear sky conditions.

Sunny Day

    

ay Cloudy Day

 

 

 

 

 

 

   

 

 

 

E ‘. . ‘ .' .' l
e 1 - j
3 _ o
I “ .41 v l
W
e
C)
C
—OTC
"'CONTHOL

 

 

 

10

 

 

 

 

 

 

.1 ’3,
Temperature “C

 

 

 

 

 

 

 

 

 

j #9 t 0 e-ﬁ ¢ r

N O N O

a: a 2 :1: . a g t
-5

TimeofDay

Figure III- I. The course of mean temperature and relative humidity in OTCS and control

plots for three different types of days: a sunny day (8/8/95); a typical day (8/6/1995); and
a cloudy day (7/30/95) (N for OTC =18; N for control = 18).

There are among-chamber differences in warming, presumably due to micro-
topography and vegetation structure. This variation is similar to among-control plot

variation but greater in magnitude. These differences are demonstrated in Figure III-2 by

50

examining the average, maximum and minimum temperature on a typical sunny day both
over control plots and within OTCS; the range in the graph represents the two extreme

monitored plots for each treatment.

111.2.2 Vertical and Horizontal Patterns of Chamber Warming

Vertical and horizontal patterns of warming were documented within the
chambers (Figures III-3). These temperature distributions are presumably due to
inﬂuences of shading, convection, and ventilation within the chambers. The north side of
the chamber is warmer because direct radiation enters through the open top when solar
radiation is greatest, around solar noon when the sun is in the south at a 42.5 degree angle
or less (Figure III-4). The vertical temperature distribution within the chambers is
notably different than the controls. Temperatures are warmer within the OTCS at all
heights and the peak in warmth is at 16 cm height whereas under natural conditions it is
warmest at the ground surface and becomes progressively cooler further from the surface

(Figure III-3).

[11.2.3 Light Distribution within 9 Chamber

Measurements of light showed that approximately 80% of the ambient solar
radiation enters the chamber, and that these levels are not evenly distributed (Figure III-
4). Due to the higher light intensity when the sun is to the south, the light levels are
greater in the northern portion (83% of ambient) of the chamber and less in the southern
portion (68% of ambient). Light levels are 93% of ambient in the center. These light

levels are considerably different than the reported transmittance of the chamber material

51

100- .. - _

‘ O;
a.‘ 8?.
\‘\‘.‘ .3; ..e
a "3:
O- ‘. ‘4 . ' .e
U ‘A o ..'
E m qr- ............................ . :‘oo'oooooo.cocoooooooooooo‘o‘co.’ ................ ov-
e‘ ‘ . OOOOOO

3 \ ' I.
I . Q - _ . a
o r"
.2 ............
on
(U
E
a; so .. ............................................................................... -_

— OTC
- - - CONTROL

 

e" ..‘
-o'. . ‘5‘... 6
‘ C ' a
O.-'..-\ .. ..‘.0.. ‘ a
..o ..e \ ..... \ . ‘
d— ooooooooooooooooooooooooooo .o 0;.“ a... ooooooooooo ‘ .9.“ :3:.‘. e‘ o ' a o oooooooooooo o-r— 5 °0
8 . ‘. O ".,
e. ‘ t ‘ ‘ m
. - ' ‘x'.\ b
,4. '. . ‘4" a
v... ‘5
‘ 8‘. . . h
‘3 0)
‘ 3

r i 1 i 1 O l2

O N ‘0

V Q (0 N N N

1- 1- N

 

 

 

Time of Day
Figure III-2. The mean (thick line) and range (thin line) course of temperature and

relative humidity in OTCs and control plots on a sunny day (818/95) (N for OTC = 18; N
for control = 18).

52

A _ u zv 32 EN :3 - cm as: see .3528
a £53 8255:. new Essie—e .52: EEO 09:26 05 he eczentaﬁ 283383 .5530: ten .39.? 2:. .2: Beam

 

 

   

   

     

  

 

can—2. Eu: _ﬂa:ONtO—.— —ﬂ°—tﬂ>
.3250 121.7 .9580 IZIV
u\__.l_ .3230
83. o. 2 83.
58a 58m .380 5.32 5.32 “M339
alllll'lllllllllllllﬁq I IILII II
552.... . e... . 2. no. . 3.. . es. _ ...3 m. .5. a... 256
5585: u n new n 9.2%. o.——@ V «.36 m 0.: Eye «.9. ad QM 3 9.... ed— —.m
\e \ 2.8: 935d ms: 31.486 x
J :39— ._.n... ”N EMF . w
as: ..-3 m." _ a: .

 

 

 

seen _. a... m."

Egan 2.. no md .53 ‘

 

 

53

of 86% in the visible wavelengths for several reasons: the StowAwayTM light loggers
used in the ﬁeld measure a different range of wavelengths that are skewed towards the
infrared (Appendix Figure A-S), the angle of the sides of the chambers to the sun does
not allow for optimum transmittances, the chambers can be dirty or scratched, and most
importantly the open top of the chambers allows direct light. Overall the chamber creates
minimal shading, due to the of the high transmittance of the chamber material and the
open top allows direct sunlight to nearly all portions of the chamber at some time during

the daily solar cycle.

 

 

     

 

a
’ ’-

North Edga

 

 

<—N—

Light Distribution

Figure III-4. Light distribution relative to natural conditions within a chamber from
August 1 -21, 1996 (N =1 and N = 2 for the center).

54

111.3 BELOW GROUND RESPONSES TO TEMPERATURE

[11.3.1 Soil Temperature

The soil temperature was only sampled during one ﬁeld season; thus the only
comparison that can be made is between treatments. The chambers caused distinct soil
warming in the wet meadow site in Barrow, Alaska. Warming was more pronounced in
the upper layers of soils and gradually diminished with depth (Figure III-5). The increase
in temperature was 1.2, 0.7, 0.4, 0.4, and 0.l°C for depths l, 5, 10, 15, and 30cm
respectively. There was considerably less soil warming under chambers in the ITEX dry

heath community than in the wet meadow community (Walker 1997, Figure III-5).

[11.3.2 Active Layer Thickness

The active layer development was signiﬁcantly different between the two years.
The warmer 1996 ﬁeld season had a thicker active layer (Figure III-6, Appendix Table A-
2). The active layer thickness did not show a statistically signiﬁcant response to chamber
warming. Plots were found to be signiﬁcantly different from each other due to overall
heterogeneity of site thaw patterns. Personal observations of differences up to 10cm in
depth of thaw between points only a meter apart were not uncommon; Mueller (1996)
also refers to such heterogenity in Banow tundra. This may mask a small chamber effect
that appears to be present at some times during the summer from an examination of

Figure III-6.

55

.2 u 75 8:8 :03 we p.82: 2: 5.23 203 53.520 92 05 E... 63— .w. 3.3.2 I c. 5...: eaten 2:: 2:3
05 ED...— 203 38:8 8.5 $25.80 03838.. :05 E... 8.5 €5.58 92 E 83820 2: .82... wig? mom 6-5 8:3"—

:33: .25 5.3... to
am... 1... a... l...
.58. 33"." ...3." n." 1.329.:
.3.—co 5%. to N... .350

 

 

       

     

        

       

.53. 5 a... N6
52. «Seed
5m. ..n 3 n .v

nnrrvdl-S ..

.5»... ...3 m."
.52. 5 3 tn

.52. Se." N."
.53. ed 3 ed
Eon. an a.» ad
. armi-n: 6.0 II I

.5»... on 3 ed
5n. on o... :6

I E M»... E.

 

.52 «3.3% .m. .5».— .en. ”a.

 

 

 

l

 

 

 

 

 

56

Progression of Thaw Depth

 

 

v 60
‘ —°—OTC
- ° -CONTROL

   

4.
01
o
5

1 996

 

................................................

oooooooooooooooooooooooooooooooooooooooooooooooo

Active Layer Thickness (cm)

................................................

 

 

 

 

Figure III-6. Progression of active layer thickness throughout the 1995 and 1996 ﬁeld
seasons (N for OTC = 24; N for control = 96).

111.4 ACCUMULATION or THAWING DEGREE DAYS

There were differences in the thawing degree day accumulation from snow melt
(TDD*) between the two years. The 1996 ﬁeld season was warmer than that of 1995 and
had a longer growing season up until late August. Thawing degree days calculated from
daily average temperatures from standard screen height measurements provided by

NOAA show the accumulations through August 18 to be 212 and 282 for 1995 and 1996

57

respectively. These two ﬁeld seasons Show a higher trend in thawing degree day
accumulation than the thirty year whole year average of 251 from 1941-1970 (Brown et
al. 1980).

The documented increase in canopy temperature within the chambers although
relatively small in absolute magnitude over the summer caused approximately a 38% and
30% increase in TDD" in years 1995 and 1996 respectively (Figure Ill-7). The among-
plot differences in temperature reported in section 11121 are ampliﬁed when examining
thawing degree days of the plots for the entire summer as seen in Figure III-7 examining
the absolute range in TDD“ per treatment. In addition the TDD“ for the 1996 controls is
similar in average and range to the 1995 open-top chambers; this will be useful for
comparing the response of plant species comparisons in Chapter IV. The thawing degree
day accumulations from June 19 -— July 6, 1995 were estimated from screen height data
provided by NOAA because data loggers had not been installed by this time, thus there

are no differences between treatments in Figure III-7.

111.5 DISCUSSION AND CONCLUSIONS

The two ﬁeld seasons showed considerably different weather conditions, which is
in keeping with the known extreme variability in daily and interannual weather patterns
for the Arctic. The average control plot temperature in 1996 of 4.8 °C was 1.3 °C warmer
than that of 1995. The snow melt of the ﬁeld site in 1996 was approximately 10 days
earlier than in 1995. The combination of these factors caused an increase of 116 thawing

degree days from snow melt (TDD*) up to August 18 in 1996 over 1995.

58

Thawing Degree Day Accumulation from Snow Melt (TDD*)

 

. . ,,-—OTc

- - -CONTROL

  
  

._-_____._____ _ _ ,_ _____________

Accumulated Thawing Degree Days

___________________________

__________________________

 

 

7-Jun i

a

3
‘F
a
.—

Day

Figure III-7. Mean (thick line) and range (thin line) degree day accumulation from day of
snow melt for the ﬁeld seasons 1995 and 1996 (N 2 7).

59

Total seasonal thawing degree day accumulation is not a useful predictor of plant
response because dormancy or senescence generally begins in mid August, thus
accumulations after this stage have little effect on the plant. For example, from personal
observation, well formed buds of Cardamine pratensis in late August of 1995 did not
flower that September; rather, they over-wintered and ﬂowered feebly in June of 1996
without setting seed. For these reasons and those given on page 38 all end-of-season
vigor measures were compared with TDD“ up until August 18 not total season thawing
degree days. Due to seredipidity, the 1996 controls had a similar degree day
accumulation to the 1995 OTCs. This created an opportunity to test the efﬁcacy of the
chambers as a temperature enhancement device. If the plants responded similarly to
degree day accumulation for both treatments, then it is reasonable to conclude that the
plants were responding to temperature and not to some other chamber effect.

The increased warming and near-ambient solar radiation levels in the northern
center of the chamber has the potential to cause greater plant response in this region of
the chamber. The temperature peak at a height of 16cm as opposed to ground surface
under the control should not noticeably disrupt the natural heat differential in plant
tissues because plant tissue temperature is highly dependent on solar radiation that
naturally has high variability (Hanson 1973). The measurements of vertical and
horizontal distribution of warming was conducted in a nearby location with less plant
canopy and some exposed bare ground; we would expect a different temperature
distribution within the wet meadow site because of the dominant plant canopy (Bay
1996). Therefore, the results are only demonstrative of the general chamber performance

and do not represent exact details of response in the wet meadow itself.

The chambers produce more variability in temperature than seen in the control
plots in both time and Space. The chambers have higher daily maxima, which are highly
dependent on sky conditions, {leading to large among-day variation. The horizontal
distribution of temperature within a chamber and the among-chamber differences are
greater than the among-control differences. Due to the greater variability of temperature
in the chambers, we would expect a greater variability of plant response within the
chambers if temperature is a major controller of plant response. Furthermore, the
increase in range of variability within chambers is commensurate with climate warming
predictions that also call for an increase in the variability of weather events (Karl er al.
1995).

The lower light levels reported in the chambers are not considered to be of great
concern to plant performance. The vast majority of experiments involving the
manipulations of light levels in the Arctic show no detectable effects of minimal shading
(Savile 1972, Chapin et al. 1995). Temperature and nutrient levels are believed to be
more limiting factors than light in this system, although light levels could become an
important factor when temperature and nutrient limitations are minimized. Yet this is
unlikely with the small levels of shading occurring within the chambers and the natural
heterogeneity in light levels due to natural canopy shading.

There were differences between the wet meadow and dry heath communities in
terms of soil warming under the chambers. The ground’s surface is warmer in the dry
heath community and there are considerable differences in soil temperatures with depth
under ambient conditions. The soil temperature proﬁles of the two sites suggest that the

soils of wet meadow conduct heat more than the soils of the dry heath and supports the

61

well-established notion that the heat diffusivity of soils is highly dependent on soil
moisture (Nelson et al. 1985). The dry heath site showed very little if any soil warming
(Walker 1997). No soil warming under a similar chamber was documented by Wookey
er al. (1993) in a dry heath site in Abisko. Wookey et al. (1993) attributed no soil
warming to be primarily due to reduced convective heat transfer because the chambers
greatly reduce wind. The dry heath community has a signiﬁcant amount of bare ground.
This exposed ground is highly receptive to wind and direct solar radiation, both Of which
are reduced by the chambers. In the wet meadow convective and radiative heat transfer
are proportionally less important due to the dense plant canopy, thus the soils in the wet
meadow are warmer than ambient under the chambers and the soils in the dry heath are
not.

The small size of the chambers presumably does not allow for increased
progression of active layer thickness, despite documented soil warming in the upper soil
layers in the wet meadow site. The chamber warming apparent in surface temperatures
under the chamber in the wet meadow diminishes with depth. Hysteresis from the mass
of permafrost around the chambers may dampen the small potential changes in active
layer thickness due to warming. Changes are also difﬁcult to detect as a result of the
heterogeneity of the permafrost (Mueller 1996). There are clear annual differences in
active layer thickness as shown in the comparison of 1995 and 1996. These difference
may be attributed to many factors including length of growing season, soil moisture,
snow cover, and winter temperature in addition to summer air temperatures (Smith 1975,

Shiklomanov 1997).

62

The changes in soil temperature and the potential for changes in active layer
thickness are of importance because of their potential to change nutrient availability and
tumover rates in the system (Hobbie 1996). Tundra systems are widely believed to be
nutrient limited, and changes in nutrient availability have been shown to cause signiﬁcant
changes in vegetation (Chapin 1987 , Jonasson 1992). The potential for the uncoupling of
below ground warming and canopy is the most notable limitation of the chambers.
Understanding the dynamics of the below ground system and its links to warming are
critical if chambers are to be used to predict species response to global warming.
Fortuitously, there are measurable increases in soil temperature in the wet meadow site at
shallow depths, were most biological activity occurs, and this warming is in accordance
with predictions of increased soil temperatures associated with climate warming.
Although this warming effect diminishes with depth and does not create a corresponding
increase in active layer thickness, it causes less concern here than it does for sites that do
not show warming, such as the dry heath site in Barrow.

There are other documented limitations of the OTCs. Marion et al. (1993)
recorded night time cooling and Jones et al. (1997) found evidence for reduced
pollination; however, neither of these phenomena have been observed at Barrow. This
thesis has attempted to address the key concerns presented by critics such as Kennedy
(1995) about the use of chambers: the chambers were installed during the snow free
period only to avoid unnatural snow accumulations and winter microclimates;
temperature and humidity was monitored throughout the entire season; and light, soil
temperature, and temperature distribution within a chamber was monitored. The most

notable limitation of the chambers is the lack of a true control for temperature because

63

they manipulate more than temperature. These secondary effects, including shading and
reduced wind, make conclusions based on plant response to chambers not solely
attributable to temperature.

These studies suggest that the use of the OTCs is valid and that they are an
appropriate tool to produce summer warming. Further, the warming levels of 1.5 - 1.9 °C
are in the range predicted by global climate simulations (Chapman and Walsh 1993,
Houghton et al. 1996). It is believed that discoveries based on short-term plant response
to chambers coupled with natural interannual variation will be of use in understanding
plant temperature relations and that long-term plant response to chambers coupled with
latitudinal comparisons will be of use in forecasting vegetation response to climate

warming.

W W
PLANT RESPONSES TO TEMPERATURE
In this study, all species contained within the site were monitored for many
phenologic and vigor measures (Tables 11-5 & II-6). The analysis presented will be
restricted to those species with sufﬁcient replicates to conﬁdently form conclusions. This
reduces the database from thirty one to thirteen species. For several of the species, the
ﬂowering variables are not discussed because of limited sample sizes resulting from the
episodic nature of arctic ﬂowering or their low frequency in the plots. Analyses were run
separately on each species, and then rerun for overall signiﬁcance for all species in a

combined data set (see section 11.4.2).

1V.l RESPONSE OF FLOWERING PHENOLOGY

Question: In what way and to what extent will the phenology of selected plants respond

to variations of growing season temperature and increased canopy temperature?

IV.1.1 Hypothesis 1a: Flowering will occur earlier under warmer conditions.

The onset of ﬂowering for all species monitored is presented in Figure IV- 1 . All
of the species ﬂowered signiﬁcantly earlier during the warmer year of 1996 than the
cooler year of 1995 and half ﬂowered signiﬁcantly earlier in the chambers during those
years. The combined overall analysis found year but not treatment differences to be
signiﬁcant. In all but two species, J. biglumis and E. triste, the average date of ﬂower

opening was earlier in chambers than the control of that season; furthermore, this

65

difference was signiﬁcant in C. pratensis, S. hirculus, L. arctica, L. confusa, H.
pauciﬂora, and D. ﬁsheri. The two species that did ﬂower on average marginally earlier
in the controls, E. triste and J. biglumis, ﬂowered earlier in the OTC’s the following year.
J. biglumis was the only species to show a signiﬁcant interaction between year and

treatment.

IV.1.2 Hypothesis 1b: Flowering win occur at approximately the some degree day
threshold.

Flowering was analyzed in terms of thawing degree days since snow melt (TDD*)
in order to examine temperature relations. If accumulated temperature is signiﬁcant in
controlling plant development than ﬂowering should occur at the same degree day
accumulation for all treatments and years rather than calendar (Julian) date. When
examining the date of ﬂower opening in terms of TDD“ rather than Julian days
interannual variability was found to be much less signiﬁcant across species (Figure IV-2).
Year was found to not be signiﬁcant overall; although S. hieracifolia, S. hirculus, J.
biglumis, H. pauciflora, and D. ﬁsheri showed year as a signiﬁcant effect on the TDD"
date of ﬂowering. Treatment generally affected ﬂowering and was found to be overall
signiﬁcant. For all species within a given year it took on average more degree days for a
ﬂowering to occur within chambers. For half of the species this effect was signiﬁcant.
Although there are often signiﬁcant chamber and year effects, the magnitude of these

effects is much less in terms of TDD“ than Julian days.

Julian Day of Flowering

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m Graminoids
mm 00......
year 0.001 U "’95
treatment 3%: I OTC 1905
mm‘ 8% D COI'II‘OI 1998
" ea .
% hears”) 0.174 I OTC 1990
o year'trea app 0.198
5 213 4 ----------------------------------------------------------- —I -----
3
3
g
‘5 _ _ _ - _
N 011 14 a
01101 0.001 0.001
P-veluee 0.001 OM

 

Speclee mm mm mm mm Guam mm WW

 

Bubs

 

 

 

 

 

 

 

 

 

 

 

Dateof Flowerhg (Julan Days)
| I
I
I
I
|
I l I
I
I
I
I
I
I I
1 1
. I I
J l I
I I
0 I |
I I
U I '
11 I I
I I
‘ 1 1
1 l
H
H l '
l I
II ' I
1 1
1
1
1
1 1
1 1
1 1
I
I
I
I
l I
|
I
I
1 1
1 1
1 1
' 1
I
1
' ‘ 913.1%», 4131; w." " 1
A 1
l

 

N 0 0 3 7 0 4 4 s 2524 2721
no no 0.

"I’M """" conveneon converter:

Spedee mm Drebehae Wm mm mm

Figure IV- 1: Julian date of ﬂowering for all species. Histogram bars represent averages,
with standard error of the mean as error bars. P-values of less than 0.05 reported from a
two way ANOVA and samples size are given for each species. Overall signiﬁcance (top
left) is reported from a three way AN OVA run on all forbs and graminoids.

67

Degree Day of Flowering

4n (Renegade

    

   
     

9099999
00
§§§§§ax

   

 

 

 

 

 

410

 

290 4

1704

Date of Flowerhg (TDD')

 

 

 

 

 

 

 

 

19 6 $12
004

 

0.
0.016
Spedee 011mm Dmbebaee Ween“ WW mm Sammie

Figure IV-2: Degree day Of ﬂowering for all species. Histogram bars represent averages,
with standard error of the mean as error bars. P-values of less than 0.05 reported from a
two way AN OVA and samples size are given for each species. Overall signiﬁcance (top
left) is reported from a three way AN OVA run on all forbs and graminoids.

68

IV.2 VIGOR RESPONSE

mm In what way and to what extent will the vigor of selected plants respond to

variations of growing season temperature and increased canopy temperature?

IV .2.1 Hypothesis 211: Plant will produce more ﬂowers under wanner conditions.

Of the six species for which the number of ﬂowers was measured there was no
consistent pattern in ﬂower numbers within the plots in response to temperature (Figure
lV-3). The three species, S. foliolosa, S. cemua and D. ﬁsheri, ﬂowered more in the
warmer year and C. stans and H. pauciﬂora signiﬁcantly ﬂowered more in the chambers.
C. pratensis and S. hirculus did not respond to either of the treatments or year. Overall
year was found to be signiﬁcant but treatment was not. Furthermore, the average number
of ﬂowers was found to be greater within the control than within chambers for D. ﬁsheri
and S. hirculus in 1996. There was considerable variation of ﬂowering between plots and
this measure was not recorded for all species; therefore statistical comparisons were of
low power.

An examination of the percentage of monitored individuals that ﬂowered also
shows no clear pattern in response to interannual variability or canopy warming (Table
IV-l). Some species ﬂowered at a higher percentage during the cooler year of 1995 and
some ﬂowered more in 1996; some species ﬂowered more in the controls and Others
ﬂowered more in the chambers. Neither year nor treatment were signiﬁcantly different in

the percentage of individuals ﬂowering, based upon the results of an AN OVA.

69

Number of Flowers

 

,1 Graminoids
I i n' 0001 [:1 Control 1995
l’ .
mg?!“ 3'03? l ore 1995
8

year'treat 0.051 D °°""°' 199°
(’22pr 0.001 I OTC 1996

' 0.004

year'trea app 0 005

U
.

________________________________________________________

Numberof Flowers per Plot

 

 

 

E E E
3 1?, 8
arm-s --------- s -------- a
2 2 2
‘6 ‘5 ‘6
Z Z Z
N o **** t """ A ‘
P-veheemam
new

Speclee .hncrrebiglumle Luzuleerctice Luzulecmfuee Encphcnmrrlete Cuexehne mm WW

51 m

U
.1
a
f
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
l
I
I
I
I
I
I
I
l
I
I
I
I
I
I
I
I
I
l
I
I
l
I
l
I
I
I
I
I
i
I
I
I
l
l
I
I
I
I
I
I
I
I
l
I
I
I
I

 

 

sured

unrrirnr‘ r
r

’0". ESL)?! ‘
iii? 3471,27... rT-i. L‘iq'u!

Number of Flowers per Plot

 

 

 

 

 

,.
n ................. 8 ................................

QI' ;‘;

° '5

z 5

£1

r:

0:-—: :—-:::;‘:::‘:: - V“

N 2424 2424 2424 2424 2317 2318 0

year 0.00 0.014

P-veluee Ire-em
Ire-rm

Speclee Cerdemhepmbnele Drebelecbe Wm Surﬁng-Halon 31mm Wham

Figure IV-3. Number of individuals ﬂowering within a plot. Histogram bars represent
averages, with standard error of the mean as error bars. P-values of less than 0.05
reported from a two way AN OVA and samples size are given for each species. Overall
signiﬁcance (top left) is reported from a three way AN OVA run on all forbs and
graminoids.

70

 

 

Table IV-l. Percentage of monitored individuals that
ﬂowered during the two years in both treatments.

 

 

_.M__ __1216_

Species Control OTC Control OTC AW
Cerdernrhe pretensls 0.0 2.8 0.0 2.9 1.5
Gemstone 13.6 15.3 20.0 25.0 18.6
Dabs Iactea 81.8 80.0 82.1 87.5 82.7
01001708 W 5.6 12.9 12.9 10.9 10.4
Erlophoaan am. Not Measured
Herodioe paucﬂcra Nor Measured
Jurors blglumb 81.3 429 61.8 50.0 58.5
Luzua arctice 46.4 43.5 38.7 69.6 48.6
Luzula comm 50.0 37.5 50.0 52.9 48.9
8am comm 0.0 0.0 4.5 10.1 3.6
Sexifraga Woes 0.0 8.9 6.1 10.4 5.8
Saxrlrags hieractole 72.7 82.8 77.1 67.7 75.0
Sadr-age him 55.9 40.0 72.2 80.0 63.0

Age 37.0 33.3 38.7 42.5 37.9

 

 

IV.2.2 Hypothesis 2b: The height of inﬂorescence will be taller under warmer
conditions.

The length of inﬂorescence was consistently longer during the warmer
ﬁeld season and within chambers (Figure 1V-4). Both year and treatment were found to
be signiﬁcant overall. All species for which there were adequate replicates but two, E.
triste and S. hirculus, signiﬁcantly responded to both treatment and year. For all species
the average height was greater in the chambers and during the warmer ﬁeld season of
1996. This is demonstrated by the step-like pattern in Figure 1V-4 the different
treatments in different years going from control 1995 to OTC 1996. This pattern is very
similar in direction and relative magnitude to the pattern of degree day accumulation for
the years and treatments (see section 111.3). Furthermore, many of the species responded

Similarly in the OTC of 1995 and controls of 1996.

71

Plant Stature

 

 

   
  
  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  
  

2. Grammotds
.001
.001
.001
.123
A .001
E .123
3 .718
g 14 «4L --------------------------------------------
0
g
'6
g 7 a» -4 — — — — — —
o 4....4
N 3021 2419 1611 1616 1511 1712 2310 1717 1212 1420
you 0.001 0.016 0.0% 0.016
P-veluee mm 0.014 0.001 01131 um
um

Species mounts mm mm Enamel-1mm Cent”

2, Baths

 

Length of Inﬂorescence (cm)

 

 

 

 

 

 

 

 

Species Card-mm Drlbebaee Wm WW Sliding-m Sending-Mum

Figure IV -4. End of season measures of the length of inﬂorescence. Histogram bars
represent averages, with standard error of the mean as error bars. P-values of less than
0.05 reported from a two way AN OVA and samples size are given for each species.
Overall signiﬁcance (top left) is reported from a three way AN OVA run on all forbs and

graminoids.

72

IV.2.3 Hypothesk 2c: The length of leaves will be longer under warmer conditions.

There was no consistent pattern in leaf length response to chambers or years
across species (Figure IV-S). Only one species, C. pratensis, signiﬁcantly responded to
chambers and year and the magnitude of response was small. Two species, S.
hieracifolia and C. stans, signiﬁcantly responded to year and not chambers. The average
diameter of rosette was measured on D. lactea and S. foliolosa; this measure is more
reﬂective of vegetative growth than leaf length. Neither species signiﬁcantly responded
to year or treatment (Figure IV-6). Overall there was no effect of treatment but there was
a signiﬁcant difference between years. The average leaf length was not always longer in
the warmer year or in the chambers.

Leaf length does not necessarily reﬂect true vegetative response because a species
may respond by increasing length or number of leaves. For a limited number of species
the number of leaves were also measured. The product of leaf length with number of
leaves was considered a more instructive measure of vegetative response although not
ideal because the length of secondary leaves could be promoted more than the length of
the primary leaf. For the product of leaf number with leaf length (vegetative measure) C.
pratensis and E. triste showed both signiﬁcant response to year and treatment while S.
hieracifolia and C. stans showed a signiﬁcant response in year only (Figure lV—6).
Overall there was a signiﬁcant difference between both years and treatments. These
results show that there is an overall vegetative response to warming but caution must be
used because it is based on a small sub-sample of the species of which many signiﬁcantly

responded in leaf length.

73

Leaf Length

 

 

 

 

 

 

 

 

 

 

 

 

 

 

n Graminoidg
9 I I n'
I' 0.010
"extent 0.517
rm 1 8%
yea a .
ar‘ 0.001
1"st 0.956
spp 0.617
E 3.1 .............................. - -
8
'5
3
3
E a.
4 >---—-8 ——————————————— 1 —— — —
2
5
Z
N 0 ‘7': ‘7'3024 3524 14162420 121 i: ‘ '37 72
0.001

 

PM“:
11W
Species mwm Lunrbuctlcl mm Eriophonmm Comm WW MW

12 FOI'bﬁ

 

on
A
V
I
I
I
I
I
I
I
I
I
I
I
|
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
|
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
|
I
I
I
_1

Length of Leaf (cm)

    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E
3
(I)
4. ............................................ 8 .....
2
‘5
Z
N o 3019 3519 7272 7272-6640 eeso'aseo 3529:;TUU
P-veluee .
Speclee Cord-mm Drab-babe Wm mm mm mm

Figure IV-5. End of season measures of the length of longest leaf. For species D. lactea
and S. foliolosa measures are the diameter of rosette. Histogram bars represent averages,
with standard error of the mean as error bars. P-values of less than 0.05 reported from a
two way AN OVA and samples size are given for each species. Overall signiﬁcance (top
left) is reported from a three way AN OVA run on all forbs and graminoids.

74

Vegetative Measure

 

 

 

 

 

 

 

 

 

 

   

 

48 Graminoids
Overall §ignm El Control 1995
"6mm §3§§ I are 1995
168 . Control
yaﬂmu 0M7 0 mm
ggwp 0M” mom mm
15, 1' 0.203
year'trea spp 0.715
2 32 1» —————————————————————————————————————————————————————————————————
E
x
8
>
3 s a a a
é '5 """ m """" 1n ““““ 07 """ ‘ ‘ ” """" 0) """
5 2 2 2 2
‘5 ‘5 ‘6 ‘5
Z Z Z Z
N 0‘3L3“““‘:C:t::£ﬁ‘; C:‘4“tf
veer 0.007
PM m 0.040

Species WW Luzrdeeredce Luau-comm Erlephonlnm Ger-arm I-h'emdrlcepeuclﬂore Droonleﬂehed

 

 

 

 

 

 

 

 

 

 

 

 

 

,, Forbg
g 32.. __________________________________________________
to
.9
x
a
i
"5
h
.3 16-
E
a
Z
N 0‘ e771 5771‘ U A 7272 7272 3430
yeer 0.001 0.001
P-veluee treetment 0.043
mum
Species Centunheprehnele Dmbelectee Warm mm mm Sexllrmehlrculus

Figure 1V-6. End of season measures of vegetative growth (length of the longest leaf
multiplied by the number of leaves). Histogram bars represent averages, with standard
error of the mean as error bars. P-values of less than 0.05 reported from a two way
AN OVA and samples size are given for each species. Overall signiﬁcance (top left) is
reported from a three way AN OVA run on all forbs and graminoids.

75

IV.3 SPECIES PATTERNS OF RESPONSE

Qgestion: Are there species responses to temperature that can be generalized or do all

species respond uniquely?

IV.3.1 Hypothesis 3a: Short-term plant phenologic response to canopy warming will
mirror responses observed in warrn years.

There was a strong similarity of signiﬁcant response to interannual variability and
canopy warming when examining the response of species ﬂowering to degree day
accumulations (Figure IV-2). Six species (J. biglumis, H. pauciﬂora. D. ﬁsher-i, D.
lactea, S. hieracifolia, and S. hirculus) responded to year and six (J. biglurnis, E. triste, H.
pauciﬂora, D. ﬁsheri, D. lactea, and S. hirculus) responded to treatment with an over lap
of four species that responded to both (Figure 1V-2). The average accumulation of
degree days until ﬂower opening was always greater in the chambers than in the control
of that season. Due to similarities in seasonal degree day accumulations it is instructive
to compare the 1995 OTCs with the 1996 controls. From this comparison only two
species, D. lactea and J. biglumis, show signiﬁcant differences between the two

populations (Table IV-2).

IV.3.2 Hypothesis 3b: Short-term response of plant vigor observed in the chambers
will mirror responses of controls observed in warm years.
The number of ﬂowers within plots showed no consistent response to chambers or

interannual variability. No species responded to both interannual variability and

76

chambers (Figure IV-3). Comparisons made on this variable are not conclusive due to

the tremendous variability among plots and the small subset of species measured.

Table IV-2. Statistical signiﬁcance reported from post factor tests of the population
means of the 1995 OTC compared against the 1996 control. The group that
responded earlier on average is also presented.

 

 

 

Degree Date Number of Length of Length of Vegetative
of Flowering Flowers Iplot Inﬂorescence Leaf Measure
S - edit year M M green
I "EL—AAA—m‘
Cam preteltele - - nsd OTC - - nsd connol nsd control
Ceroxm nsd OTC nsd OTC nsd OTC 1'" control " cooool
Drlba lame ” control - - nsd OTC 118d OTC - -
Drporrtr’e W nsd cannot " control nsd OTC nsd OTC nsd OTC
Ellophorun Meta nsd control - - nsd OTC nsd OTC nsd OTC
Hierochloe pemﬂora nsd control - - nsd OTC - - - -
Jurcus biplane 1" control - - nsd control - - - -
Lula erotica nsd OTC - - nsd OTC nsd OTC - -
Luzula confuse nsd control - - ” OTC nsd OTC - -
Sexifra cemue - - 1‘ 1' control nsd OTC nsd control nsd OTC
Sexlfraga Woes - - nsd control nsd control nsd OTC - -
Saxifmga New nsd control - - our control nsd control "I control
W m nsd OTC nsd 000901 nsd OTC - - I - -

 

 

 

 

 

 

nsdnostatistlealdiﬂerence ”statisticaldltlerence

E.
i
s

The height of inﬂorescence responded signiﬁcantly to both years and chambers.
S. hirculus was the only species to respond differently to years and treatments.
Fortuitously, a comparison of response to interannual variability in temperature and
chamber manipulation of temperature can be made due to the fact that similar amount of
degree day accumulation from snow melt (TDD*) during the 1995 OTC and 1996
control. For all but two species, L confuse and S. cemua, the height of inﬂorescence was
not signiﬁcantly different between the two populations.

There was no clear pattern of vegetative response to canopy warming or
interannual variability; therefore a comparison is of little value. Although, of the four

species that did respond to year, two of them, E. triste and C. pratensis, also responded to

77

treatment. The two species that only responded to year, C. stans and S. hieracifolia, may
have been responding to climatic features other than temperature. The response in
vegetative growth was inconsistent for treatments and years; furthermore, E. triste

responded positively in chambers and negatively in the warmer year.

Table IV-3. Summary table of signiﬁcance of year and treatment of all species. All
responses were positive in relation to warmth (i.e., earlier ﬂowering; greater numbers
of ﬂowers, length of inﬂorescence, length of leaf, and vegetative measure) except for
E. m'ste which had a decreased vegetative response in the warmer year (bold).

 

 

 

 

 

 

 

 

 

Julian Date Number of Length of Length of Vegetative
of Flowering Flowers Iplot Inﬂorescence Leaf Measure
Species Year Treat Year Treat Year Treat Year Treat Year Treat
Cardenﬁne pretends - - nsd nsd - - u as u as
Cam m it t. nsd .0 t. it t. nsd it nsd
by.“ m 0* 08d - , it it nsd nsd _ _ I
W W 0* it it nsd I! it nsd nsd nsd nsd
Eliophorum triete "1' nsd - - nsd nsd nsd nsd are n
W W as use , as an an - _ - - |
Jm W #0 nsd _ _ it it - , _ - I
m m u as _ _ as as nsd nsd - _ i
I "an mg. ee ea ,_ _ as ea nsd nsd - ' I
Sam corms - - nsd nsd " 1" nsd nsd nsd nsd I
Saxlfraga folioloea - - ee nsd as ea nsd nsd , -
my, mm as 118d - - use are as nsd es nsd I
W him " ” nsd nsd “ nsd - - - - I
- no comparison possible nsd no statistical difference ” statistical difference

IV.3.3 Hypothesis 4: Plant species will respond individualistically to temperature.
All species signiﬁcantly responded to year or treatment for at least one plant
measure recorded (Table IV-3). No two species responded similarly to all variables
measured. For all variables measured, species were signiﬁcantly different (Table IV-3).
This does not signify that species respond differently but rather for each variable some

averages were different across species. For all measures except the degree date of

78

ﬂowering species responded differently to year as expressed by the species year
interaction (Table IV-4). Species only responded signiﬁcantly different to treatment for
the number of ﬂowers per plot. This indicates that the trends of number of ﬂowers per
plot in response to chambers were different between species, i.e. some species responded

with more ﬂowers and some species responded with fewer ﬂowers.

Table IV-4. Summary table of P-values from overall tests for all measures presented.

 

 

Effect Julian Date Degree Date 33:; Length of Length of Vegetative
of Flowerrng of ﬂowering Plot Inﬂorescence Leaf Measure
Species 0.001 0.001 0.001 0.001 0.001 0.001
Year‘species 0.003 0.900 0.001 0.001 0.001 0.001
Treat‘species 0.174 0.230 0.004 0.123 0.956 0.203

 

 

IV.4 DISCUSSION AND CONCLUSIONS

IV.4.l Hypothesis 1: Response of Flowering Phenology

The Julian day of ﬂowering was statistically different between years for every
species presented (Table IV-2, Figure IV-l). The 1996 ﬁeld season was much warmer
than that of 1995 as shown by the differences in average temperature, snowmelt, and
degree day accumulations (see section 111.1) and this warmth was likely to have
inﬂuenced ﬂowering. The date of snow melt and degree day accumulation has been
shown to effect onset of ﬂowering in other systems (Leith 1974, Walker et al. 1995,
Walker 1997). Nearly every species ﬂowered earlier on average within the chambers,

and several species, C. pratensis, S. foliolosa, and S. cemua, only ﬂowered within the

79

chambers during the cooler ﬁeld season of 1995 (Figure IV-l, Table IV-2). These
ﬁndings support the assertion that temperature controls the phenology of ﬂowering.

By expressing the onset of ﬂowering in terms of thawing degree days from snow
melt (TDD*) rather than Julian days, a more direct assessment of temperature effects can
be determined (Figure IV-2). If the only effects of interannual variability and chambers
were a change in temperature, then neither should have a signiﬁcant effect on ﬂowering
when expressed in terms of degree days. The onset of ﬂowering for a species should
occur at the same degree day accumulation if temperature is the only controller of
ﬂowering. L arctica, L conﬁrm, and C. stans ﬂowered at nearly the same TDD“; thus
for these species ﬂowering is highly temperature dependent. Other species (such as D.
lactea, J. biglumis, E. triste, and P. arctica) ﬂowered at a different TDD“ for each
treatment and year. Most of the species fall somewhere in between the two extremes,
implying that ﬂowering is controlled by a combination of temperature and other factors.
Because species have their own unique evolutionary histories, they are expected to
respond individualistically (Gleason 1926, Sorensen 1941). Species will vary in the
plasticity of their response to temperature; by deﬁnition it is expected that aperiodic
species (Sorensen 1941) will respond more in short-term experiments than periodic
species (see section 1.2.3-l). Walker (1997) reported similar individuality in response to
temperature for the Banow plant species.

It is presumably ecologically and evolutionarily advantageous for a species to
reach the same phenophase at approximately the same Julian date regardless of when
snow melt occurred or how many degree days accumulated, given the distinct differences

in weather variability throughout the season (Myers & Pitelka 1979). For example, some

80

species have been shown to modify their rate of phenologic development following
increased snow pack accumulations (Johnson 1969, Webber er al. 1976). Moreover,
plants in Barrow generally enter dormancy or begin senescence in early August although
average temperatures remain above freezing well into September (Miller et al. 1980).
This is seemingly peculiar, especially given the extremely short nature of the growing
season, and is most likely an adaptation to overcome the extreme variability and
unpredictability of weather in August (Myers & Pitelka 1979). Species may have
signiﬁcantly responded to year and treatment in terms of degree date of ﬂowering
because they are in synchrony with environmental and biological factors which are not
simulated by the chambers; for example, day length and pollinator activity. For species
that responded signiﬁcantly to treatment or year the magnitude of response was generally
less in terms of TDD“ than Julian dates. This provides further evidence that temperature
is an important determinant of the onset of ﬂowering.

In summary, these data support the notion that the onset of ﬂowering is
determined by temperature (Table IV-S). The ﬂowering of some species is more closely
linked to temperature than are others, and there is a wide variety of response; however,

all species show some correspondence between ﬂowering and temperature.

IV.4.2 Hypothesis 2: Vigor Response

Flower buds are preformed at least one year in advance in nearly all arctic
species; therefore, the ﬂowering in a given year is highly dependent on the climate of a
previous year or years and a clear response in ﬂowering to a current season’s temperature

regime is not expected (Sorensen 1941). Furthermore, large variation in ﬂowering

81

between plots made differences in treatment or year hard to determine. Thus, multi-year
bud formation times and heterogeneity in ﬂowering might be invoked to explain the lack
of detectable response in the number of individuals ﬂowering within the chambers
(Figure lV-3, Table IV -3). In order to evaluate the ﬂowering response without
confounded differences in species abundance among plots, the percentage ﬂowering was
calculated (Table IV-l). From this data no clear patterns emerge. Changes in ﬂowering
percentage and consequently abundance are predicted to occur as a response to chamber
canopy warming, but were not found. Subsequent years of this study will be needed to
properly test this hypothesis because of multi-year bud formation times. In the short-term

there was no generalizable response in the numbers of ﬂowers within a plot to

temperature.

Table IV-S. Summary table of the number of species with
signiﬁcant response in their phenology of ﬂower opening
(number responding I number examined).

 

 

 

 

Year Treatment it in common
Julian Day 10/10 5/10 5
Degree Day 5 / 10 6/10 4
# in common 5 3

 

 

 

 

' species not included: S. cemua. C. pratensr’s e2 S. foliolosa.

The highly signiﬁcant effect of both treatment and year on the height of
inﬂorescence strongly suggests that the plant’s stature is directly or indirectly controlled
by temperature (Figure IV-4, Table IV-3). This increase in length of inﬂorescence may

be a result of an adaptation to allocate more resources to reproduction during favorable

82

conditions. Growth and elongation of inﬂorescence is not predetermined as the actual
bud formation is, therefore the inﬂorescence can directly respond to a current season’s
climate, although it may be inﬂuenced by previous years stored reserves. Increased
height may be a result of a balance between short stature to maximize tissue temperatures
and the selective advantage of tall stature to raise seeds above the plant canopy to
increase potential dispersal distances (Warren Wilson 1957, Savile 1972). The dogma
among tundra ecologists is that short stature is an adaptation to maximize tissue
temperatures (e.g., Bliss 1988). By maintaining short stature higher metabolic rates can
speed fruit development because the fruiting body is closer to the warm ground surface.
As a plant increases its height, it rises further above the ground and it is subjected to a
harsher environment. Many species have evolved other mechanisms to maintain high
tissue temperatures, such as ﬂower shape and dark pigmentation (Bliss 1962, Molgaard
1982). Within the chambers and in 1996 the temperature at 130m height was warmer and
could have allowed individuals to achieve higher stature while maintaining high tissue
temperatures (Table HI-l). An example of the importance of both tissue temperature and
the height of inﬂorescence for dispersal is Dryas which has evolved to mature fruits near
the ground’s surface and then expand the pedicle after fruit formation to raise the mature
seeds and increase potential dispersal distance. If the inﬂorescence is above the ﬁrst
covering of snow, seeds may be released above the snow layer and be blown by the wind
for hundreds of miles during the long winter and in doing so greatly enhances dispersal
distance (Savile 1972).

It has also been proposed that short stature is an adaptation to allocate more

energy into seed development that would otherwise be allocated to the inﬂorescence

83

(Savile 1972). With increased photosynthate produced under more favorable
microclimatic conditions, limited allocation to stature would not be needed.

Regardless of the adaptive reason for the temperature and height of inﬂorescence
relationship there is a clear correspondence for nearly all species in the wet meadow
community monitored in Banow and the dry heath community (Walker 1997). The
increase in stature could be the result of cell elongation or increased cell number. This
distinction could be very important in relation to nutrient limitations. If the increase were
due to cell elongation, then the species would not be as dependent on nutrient availability

and a species would be more able to respond under nutrient limitations.

Table IV-6. Summary table of the number of species with signiﬁcant
responses in height of inﬂorescence and number of ﬂowers (number

 

 

 

 

 

 

 

 

responding I number examined).
7 Year Treatment it in common
Inﬂorescence Height 10/ 12 11 I12 10
Number of Flowers 3 I6 2 I 7 0
# in common 3 2
species not included:

C. pratensis (Inﬂorescence Height)
D. lacrea. S. hieracifolia. J. biglumis, L arctica. L. conﬁrsa. E. trr'sre. & H. pauaﬂom (Year)
D. lactea, S. hieracrfoll‘a. J. biglunu's, L arctica. L conﬁlsa & E. triste (Treatment)

The response in vegetative growth to temperature was variable. Several species
responded to year but not treatment and not all responses were positive (Figures IV -5 &
IV-6, Table IV -4). The signiﬁcant negative effect of year may be related to other

climatic parameters such as precipitation. The lack of response may be a result of

predeterminate or periodic growth of species that is not plastic (see section L2.3-l);
furthermore, for some species growth may have been advanced early in the season but
not late in the season. This was found in the dry heath community in Barrow (Walker
1997), but was not detectable with end of season measures. Transplants of species into
warmer microclimates have shown that the larger size of the more southerly individuals
is Often genetically predetermined (Shaver et al. 1986). The lack of response may also be
due to nutrient limitations. In an environment with low nutrients and no light limitation
leaf elongation is not necessarily beneﬁcial and it may be more advantageous to allocate
additional resources sequestered in a more favorable microclimate to roots or
reproduction. The overall signiﬁcant effects of year and treatment reported from the
vegetative measure suggest that the community is responding vegetatively but this is
most likely a result of the selection of a subset of positively responding species in the
analysis as seen in Table lV-4. It is also possible, although highly unlikely, that the
mixing of many different aged individuals has masked a response. The studied species
change in size over year as they age, particularly the rosette species; therefore a more
reﬂective measure of response may be the difference between years. This measurement
can be used as the experiment continues because the same individuals will be measured

in consecutive years.

IV.4.3 Hypothesis 3: Chamber Efficacy
The observation that it takes on average more degree days for the onset of
ﬂowering implies that there were unnatural chamber effects and that they do not act as

perfect surrogates of interannual temperature variation (Figure IV-2). Although, the

85

Table IV-7. Summary table of the number of species that showed
signiﬁcant vegetative response (number responding I number examined).

 

 

 

 

Year Treaunent # in common
Leaf Length 2/ 10 1 I 10 1
Vegetative 4 I 6 2 I 6 2
# in common 2 l

 

 

 

 

species not included:
S. hirculus, J. biglurnis. J: H. pauciﬂom (LedLength)
D. lactea. S. foliolosa. S. hireullrs, J. biglumis. L arctica. L conﬁrsa & H. pauciﬂora (Vegetative)

response of the chambers is very similar to that of interannual variability when examining
the number of signiﬁcant responses (Table 1V-2). This could be due to a balance
between warmth and developmental time requirements. Degree days do not account for
developmental time requirements as directly as Julian days. This lag may also be due to
biological cues in the plants as presented in Section IV.4.1. The consistency of higher
TDD" accumulations for ﬂowering within chambers suggests that it is not due to plant
biology but some attribute of the chambers. Higher TDD'I' accumulations for phenologic
progression was also reported for the dry heath community in Barrow (Walker 1997).
The active layer thickness in 1996 suggests that the soil temperatures were warmer and
that a larger nutrient pool was available than in 1995. Other studies have shown a strong
interaction and even synergism between temperature and nutrient availability (Parsons et
al. 1994, Chapin er al. 1995). Therefore, the warming effect of chambers may be
different than natural interannual variability because, in the chambers, the soil
temperatures may be uncoupled from air temperatures. This would cause differences in

nutrient availability between warming associated with interannual variability or

86

chambers. This interaction is a very plausible explanation for some of the differences
between interannual temperature variation and chamber canopy warming seen in this data
set. For example, in the comparison of the 1995 OTC population with that of the 1996
control, whenever there was a signiﬁcant difference the control population always
responded more except for L confusa (Table IV-2).

The increase in daily range of chambers temperatures towards the maxima may
alter the quality of warmth in the chambers. This could also explain differences between
degree day thresholds for the onset of ﬂowering.

lnterannual differences could be attributed to major differences in snow melt as
well as other climatic parameters such as precipitation which the chambers are not
deliberately designed to modify. Even with the limitations discussed above the overall
response in ﬂowering phenology to interannual variation and chambers is similar and
likely a result of temperature control. The similarity is conﬁrmed by a comparison of the
1995 OTC and 1996 OTC from which only two species, D. lactea and J. biglumis, had
signiﬁcant differences in ﬂowering in terms of TDD“ (Table IV-2).

Among the vigor characteristics measured there were no results that strongly
suggest that the species were not responding similarly to the warmer canopy associated
with interannual variability and chambers. In particular, height of inﬂorescence data
suggest the chamber and interannual response are the same; although there does not seem
to be a clear correspondence between temperature and vegetative response. The
differences in leaf length were only signiﬁcantly different between years or treatments for
two species: C. stans which was longer in the warmer year and C. pratensis which

responded to both year and treatment. C. prarensis was the only species to show a

87

consistent increase in leaf length to temperature enhancement caused by chambers and
interannual variability.

The ﬁnding that the variability in temperature was greater within the chambers
leads to a secondary hypothesis that if the species are responding to temperature the
variability of response should also be more variable within the chambers. This
hypothesis was tested by examining the residual error of an AN OVA of length of
inﬂorescence performed on both treatments separately. Length of inﬂorescence was
chosen because it was the variable that responded most clearly to temperature. The
residual errors were signiﬁcantly greater within the chambers (6.4 and 11.4 for the
control and OTC respectively). The greater variability within the OTCs is also apparent
from an examination of the error bars of all the graphs presented. This ﬁnding supports
the secondary hypothesis that there is greater variability within the chambers.

Species were found to respond similarly to the effects of chambers mid
interannual variability in temperature. 70% of the time a species responded similarly to
both year and treatment for all measures recorded. If a species responded to either year
or treatment then 58% of the time the species responded to both. Furthermore, from a
comparison of the populations of the 1995 chambers with the 1996 controls 80% of the
time there was no statistical difference for all measures presented. The most logical
conclusion from these ﬁndings is that the species are responding directly or indirectly to
temperature. From this it is reasonable to conclude that the short-term response to
chambers is a useful forecaster of response to natural interannual variability in
temperature. Therefore, it is reasonable to conclude that the long-term plant response to

chambers will be a reasonable predictor of species response to global warming, although

88

caution must be used because chambers are not perfect surrogates of natural variation.
Results gained from chambers should be used in conjunction with information on
latitudinal and spatial gradient comparisons and plant biology. The combination of these
results is necessary to form fundamental understandings of species temperature relations,
which can subsequently be used for preliminary forecasts of species response to global

warming.

IV.4.4 Hypothesis 4: Individualistic Nature of Species

Plant responses show a clear individualistic response to temperature enhancement.
No two species responded similarly to all measures recorded and all species responded
signiﬁcantly to at least one character. Similar results have been reported from many
studies and this result is not surprising (Chapin & Shaver 1985b, Oberbauer et al. 1986,
Walker 1997). Nevertheless the species in the wet meadow responded in a similar
fashion in some characters. For example, all but two species, E. triste and S. hirculus,
signiﬁcantly increased their length of inﬂorescence during both the warmer ﬁeld season
and within the chambers.

It is clear from an examination of all species that species have different collective
adaptive strategies in response to temperature and allocation patterns vary among species.
The response of species in vegetative growth and stature will change canopy structure.
These changes in addition to others, for example root response, are likely to alter the
competitive ability of a species for resources and this will lead to changes in community

composition and abundance. Therefore, it is predicted that the long-term response to

89

warming will be signiﬁcantly different than the observed short-term response due to

species interactions (Chapin et al. 1995, Molau 1997).

IV .4.5 Summary and Conclusions

The major conclusions of plant responses to warming in the wet meadow are
based on a synthetic average species response. Plants were found to respond similarly to
the chambers and interannual variability 70% of the time. These similarities in response
show that the chambers are useful simulators of interannual variability in relation to
temperature. Nevertheless, the response of ﬂowering phenology in terms of TDD“ was
that more species responded to treatment than to year, and the average species ﬂowering
was always after a greater accumulation of degree days in the OTC’s than that season’s
control. For all other characters the response to year was generally greater; this was
likely due to more differences between years than temperature. It is also possible that the
air and soil temperatures within the chambers are not strongly coupled as suggested by
the active layer data for the two seasons. This notion would provide a plausible
explanation for the reduced magnitude of response associated with chamber warrrring. It
is possible that plants in the chambers are more nutrient limited due to cooler soils than
when they experience the same level of warmth in a warm year. It is also possible that
the quality of warming within chambers is not comparable to naturally warmer conditions
because of the increase in daily range of temperatures. Therefore, predictions of species
response to temperature must be cautiously made from a combination Of results from
chambers and warmer years only if they are supported by fundamental biology.

Nevertheless, due to the strong similarity in species response to short-term chamber

warming and interannual variability, it is reasonable to conclude that long-term plant
response to chambers will be analogous to plant response to global warming.

The onset of ﬂowering for species in the wet meadow responded greedy to
temperature, yet the numbers of ﬂowers did not. The likely explanation for no change in
the numbers of ﬂowers is that this short-term experiment did not encompass the multi-
year bud formation times required for most arctic species. The vegetative responses were
found to be variable between species. This is presumably due to the differences in
growth strategies (aperiodic and periodic) and likely related to interactions with abiotic
and biotic attributes other than temperature including nutrient limitations.

The most consistent plant response was an increase in the length of inﬂorescence
in warmer microclimates for nearly all species. Arctic species allocate larger proportions
of energy into sexual reproduction than their temperate homologs, although vegetative
reproduction is prevalent and perceived as the norm. For example, ﬂowers of arctic
species are generally disproportionately larger than temperate counterparts (Savile 1972)
and the allocation of energy to produce seeds is perceived to be higher in arctic species
(Chester & Shaver 1982). In fact the effort measured in carbon cost of producing a new
tiller of Eriophorum by sexual reproduction has been measured to be 10,000 times greater
then propagation by vegetative reproduction (Chapin & Shaver 1985a). This suggests
that the beneﬁts of sexual reproduction must be great (Bliss 1988). Because of the
implied beneﬁts of sexual reproduction and its infrequency it is expected that plants at
Barrow will allocate more energy to reproductive characters rather than vegetative

characters during more favorable conditions. The consistent increased length of

91

inﬂorescence during warmer conditions suggests that plant species are able to capitalize
on warm opportunities by allocating more energy to reproduction.

Arctic plants are in a perpetual trade-off to maintain low stature in order to
maintain high tissue temperatures for development, photosynthesis, and translocation and
to increase stature to raise seeds above the plant canopy in order to enhance competition;
for example, dispersal potential and pollination. With increased canopy temperature the
balance can be shifted towards taller stature. If this increased stature were due to cell
elongation it would be less dependent on nutrient availability and able to respond more
than characters that are nutrient demanding. This could become important with predicted
changes in nutrient availability, turnover, and demand predicted by warming (Chapin et
al. 1995, Hobbie 1996).

The combined data show different allocation patterns for various species in the
wet meadow. These differences are likely to result in changes in species competitive
abilities and ultimately lead to changes in community composition and abundance due to
species interactions. Therefore it is predicted that the short-term proximate effects of
warming presented in this thesis, which are nearly all positive, will be very different than
the long-term ultimate effects of warming. From this the predicted ultimate effect of

warming is changes in community composition, structure, and function.

92

Chapter V
CONCLUDING REMARKS

v.1 RELEVANCE OF THE STUDY IN RELATION TO CLIMATE CHANGE

Arctic summer temperatures are expected to increase by about 4 °C over the next
50 years (Dickinson & Cicerone 1986, Houghton er al. 1996) and have recently been
rising at the rate of 0.75 °C per decade (Chapman & Walsh 1993). These changes have
important feedback implications via the effect that vegetation has on atmospheric
processes (Shaver & Kummerow 1992, Henderson-Sellars & McGufﬁe 1995). Yet there
is insufﬁcient information on the direction and the rate of change due to direct and
indirect responses of the arctic ﬂora, vegetation and soils to warming to make reliable
predictions of the future (Chapin 1984, Robinson & Wookey 1997). This project has
contributed, and will continue to contribute, to the growmg body of knowledge on tundra
plant responses to experimental warming (e.g., Chapin er al. 1992, Callaghan et al. 1995,
Henry & Molau 1997, Oechel er al. 1997). A large, continuous database is essential for a
comprehensive understanding of the system since responses vary widely with latitude and
habitat, and long-term responses have generally been found to be different from short-
terrn responses (Chapin et al. 1995, Molau 1997). Further, changes in the arctic climate
will vary by geographic region (Chapman & Walsh 1993), and will become more
variable and extreme ( Karl er al. 1995, Houghton er al. 1996). Tundra plants respond in
highly individualistic ways to a range of environmental factors including temperature
(Webber 1971, Chapin & Shaver 1985b, Chapin er al. 1996). For these reasons, the
International Tundra Experiment (ITEX) was established to study both direct and indirect

biotic changes to changing climatic factors over a long enough period to encompass

93

extreme events in order to predict future vegetation and plant patterns over the entire
Arctic (see section 1.3.2).

The wide taxonomic breath of this study combined with the detailed
documentation of chamber performance makes this study the most extensive in ITEX.
This study has already contributed a substantial proportion of information to the ITEX
network and has a sizable inﬂuence on conclusions of the group (Arft et al. in prep). It
has also made a strong contribution in understanding the chamber dynamics and their

weaknesses and strengths as forecasters of species responses to warming.

v.2 SYNTHESIS OF THE SYSTEM RESPONSE

Nearly every species responded to warming by increasing its stature in the wet
meadow. There was no consistent response among species in terms of leaf length.
Although there was an overall increase in vegetative response including leaf length and
number, this increase was not consistent across species and may be an artifact of the
small subset of species used in the analysis. The consistent increases in stature indicate
that an investment in current season’s reproductive effort is a short-term response to
increased temperature. The magnitude and occasionally the direction of response was
variable across species for vegetative characters. This indicates that there is a potential
for changes in species composition and abundance. Other studies have shown differences
between short-term and long-term responses to temperature enhancement (Chapin er al.

1995, Molau 1997). These differences have been strongly tied to nutrient availability and

species competition. It is predicted that these differences will also occur in the wet
meadow community in Barrow.

The implications of increased stature in almost all species, and increased lengths
and numbers of leaves in some dominant species, lends support to the prediction that
communities such as the wet meadow in Barrow will respond to warming with an
increase in canopy height and leaf area index. This could change the surface energy
exchange process that could further inﬂuence regional temperature (Chapin et al. 1997).
A thicker mat of vegetation may also act to insulate the air above the canopy from the
soils (Tyrtikov 1959, Shiklomanov 1997). If so, as a result of plant response the soil
temperatures may not warm proportionally and may actually cool (Brown & Andrews
1982). This could lead to great nutrient limitations in the soils and shift the selective
advantage from aperiodic species that respond greatly to temperature increases to species
that are more efﬁcient in nutrient adsorption. The decomposition rates of the system also
have very important feedback links to the global system because of the large mass of
stored carbon in the tundra soils (Anderson 1991, Hobbie 1996, Waelbroeck er al. 1997).

Changes in plant canopy could effect all aspects of the community. With
increased canopy height and leaf area there will be more cover for lemmings, and it may
be predicted that their numbers will increase. However, animal population cycles are
complex and a realistic prediction is not available (J efferies er al. 1992, Chitty 1996).
Furthermore, with increased nutrient limitations food quality may decrease (Bryant &
Reichardt 1992). Accurate system-level responses to warming are currently unobtainable

given the present understanding of the dynamics of biotic responses to warming and

95

feedback processes. Identifying and understanding similar community level dynamics is

a diver of this ongoing research program.

v.3 IMPLICATIONS OF THE STUDY

A clear conclusion of this study is that species in the wet meadow community in
Barrow respond to temperature increase. Equally important is the similarity in species
responses to interannual variability and chamber warming. The chambers were shown to
reasonably simulate natural variations in temperature; they increased canopy temperature
and caused some soil warming. From a detailed documentation of the chambers over the
two ﬁeld seasons it is clear that the chambers warm the canopy in a predictable manner
and at a magnitude that is commensurate with forecasted changes in regional climate. It
is also clear that the responses of species are similar to both a warmer canopy caused by
chambers or warmer years. This ﬁnding suggests that the species are responding, either
directly or indirectly, to temperature. However, the chambers may have potential
limitations due most notably to potential uncoupling of soil and canopy temperatures.
Therefore, conclusions based on chambers must be made in conjunction with other
ﬁndings. Furthermore, strictly empirical predictions of short-term response to global
warming based on chambers will be insufﬁcient and predictions must be based on a
combination of response to chambers and natural interannual variation. Long-term
response to chambers and interannual variation will not be analogous due to cumulative
effects of warming on prior seasons. For the latter predictions a fundamental

understanding of process is essential. Only from a combination of processes based on

96

multiple experimental procedures can realistic predictions of plant response to warming
be made.

V.4 FUTURE WORK

The results presented in this thesis are part of an ongoing research program on the
effects of temperature conducted on a wet meadow and dry heath community in Barrow
and at a complementary wet meadow and dry heath community approximately 100 km
south in Atqasuk. The study will continue to use the basic experimental design and
OTCs. In total there are four primary ﬁeld sites, each with 24 chambers and 24 control
plots. From this large array we are able to examine a large number of species responses
and examine how these responses vary among communities and over climatic gradients.
In order to make comparisons across diverse taxa, we have been exploring similarities in
plant functional type response in order to determine the best predictors of species
response.

These manipulations will be continued over many years in order to compare
distinctions between short-term response and long-term response. It is hoped that these
longitudinal experiments will encounter extreme events that have been shown to often be
critical to community composition (Grime 1990, Sparks & Carey 1995). The predictions
based on these long-term manipulations will also be compared with historical plots
sampled more than twenty years ago. From this we will determine if there have been

changes in species composition and if any of the potentially Observed changes are

97

commensurate with predictions based on the observed warming trend of the last twenty
years.

In addition, small-scale experiments have been run to determine effects of wind,
chamber size, biomass, and revegetation. Many more speciﬁcally focused experiments
are planned with a particular emphasis on the below ground response to warming. The
goal of these experiments will be to determine process and ﬁll the gaps in our
understanding of the system’s dynamics (Table V—l). For a conceptual diagram of the

future research program see Figure V-l.

Table V- l . Three potential small-scale experiments speciﬁcally designed
to test hypotheses presented within this thesis.

 

- A detailed documentation of plant tissue temperatures throughout the summer that will
establish direct chamber effects on the plants.

0 A correlation of soil temperatures at various depths with canopy temperatures to
determine if there is buffering of soil temperatures under chambers.

0 Cell counts of inﬂorescences from selected species to determine if increased stature is a
result of elongation or increased cell numbers.

The goal of the future work is to integrate and combine our results into a new
synthesis that will have application in predicting the response of the Arctic System to
global climate change. The project will strive for integration between the paradigms of
the individualistic species and of plant functional types. The state of knowledge about
arctic plant ecology and physiology, combined with limitations of modeling methods,
preclude using the species as a basis for modeling future arctic vegetation. Therefore,
modelers seek to predict vegetation change based on a limited number of taxa such as
functional groups or plant life forms (Soloman & Shugart 1993, Woodward & Cramer

1996). The future work will examine plant responses to warming within various

98

 

A Study System

QT.1Short-term Res nse 0T2 Communi Res nse
[ £2.._\ r——‘——'L_&‘ﬁ

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Q—Wm
.
k j k m j
QT.3 Plant-Soil Fles nse
"W
L J
B OT.4 Temporal and Spatial Dimensions
4 [
f r
l
l
h
l
A
12:: ESE]:
We: :21 1:: 1:1 1:1
Meadow I: E I: 1:)
m C:I[:ﬂ:_—J (2::
E
3 (:52: 1:13:11:
0., 1:) 1:1 1:: 1::
Heath 1:] 1:1 I: 1::

 

 

 

 

 

 

 

 

 

 

 

 

 

TEMPERATURE

 

Figure V—l. (A) A conceptual diagram of the tundra plant-soil system components
predicted to respond to temperature change and monitored by this ongoing project. The
principal domains of the Question Themes (QT) and their regions of overlap indicate
important system linkages. (B) The diagrams in A are extended to depict the spatial and
temporal dimensions of the project.

traditional species groups based on functional type or evolutionary history and create new
“temperature response groups” based on species response to warming (Catovsky 1998).
This classiﬁcation will be created from empirical knowledge of species response to
interannual variation and chambers combined with fundamental understandings in
biology with emphasis in plant processes and genetics. From this the project will strive
towards reasonable predictions of species response to warming in order to provide
information to those researchers attempting to realistically model future changes in
species composition and abundance due to future climatic warming (e. g., Prentice er al.

1992, Woodward & Cramer 1996, Diaz & Cabido 1997).

100

APPENDD(

101

APPENDIX

Internal temperature logger with optional external probe

SiowAway'“ XTl

Features

0 External probe measures three ranges: -5°C to +37°C, -37°C to 146°C,
-39°C to +122°C

Internal sensor overridden by external sensor when plugged in
Standard external thermistor probe lengths ofl, 2 and 6 feet

The year battery life (user replaceable)

Nonvolatile EEPROM memory retains data even when battery is removed
Safe operating temperature range of logger is -39°C to +76°C

Small size: 1.8" wide x 1.9" tall x 0.6” thick and 0.91 oz.

Optional submersible case rated to a 400' depth

2K, 8K or 32K memory sizes, storing 1800, 7944 or 32,520 monuments
Start the logger, readout and plot the data with LogBook‘ software

for Windows or Mac

42 preselected intervals from 0.5 second to 4.8 hours, corresponding to
deployment durations up to two years

Blinking light conﬁrms operation

Alarm indication

Programmable delayed start (up to three months)

Multiple sampling with minimum, maximum

or averaging F
0 Push button triggered start

0 Data exportable to spreadsheet programs (lotus, Excel, etc.)

0 DOS and batch utilities available

Overview

The StowAway“ 111'! is a miniature, reliable temperature logger which operates with LogBook‘ soa-
ware for PCs or Mace to produce timeItemper-ature data. The StawAway XTI is equipped with an
internal eensorformonitoringairtemperatures. Usingtheoptionalexternal thermistoreableenablee
accurate measurements in water or hard-to-reach places.

lndivldual callbratlon

Onset Computer Corporation’s proprietary test procedures eﬂectdvely
eliminate the resistor and AD errors, leaving only the thermistor accu-
racy, quantization error, and a small residual calibration error. Plot A at
therightshowstheworstcaseerrorfcrthethreestandardtemperature
ranges of the StowAway XTI.

 

Temperature resolution

Plot B shows the temperature resolution of the StowAway XTI for the
three temperature ranges. The resolution is the diﬂ‘erence between
adjacent temperature steps that. the logger can record.

Thermal time constant

The thermal time constant (90% response to a step change in tempera-
ture) of the external thermistor sensor, used with the StowAway XTI, is
less than ﬁfteen seconds when it is in stirred water and less than three
minutes in air. The StowAway XT'I's internal sensor shows less than a
ﬁfteen minute time constant in air.

@HSQI Tel: (503) 583.9000 0 Fox: (508) mom 4 ass: (508) 563-2269 0 email: salesOonsetcorrpcom
536MacArthurBivd. 9 BMW 9 Pocosset. 11471025593450

Figure A- 1. Speciﬁcation Of the StowAwayTM temperature data logger manufactured by
Onset Inc.

102

APPENDIX

Temperature logger with external temperature probe

HOBO’XT

Features

0 Externalprobemeasuresthreestandardranges:-5°C
to +37°C, -37°C to 446°C, -39°C to +123°C

0 Tho year battery life (user-replaceable)

0 Safe operating temperature range of logger is
«39°C to +75°C, non-condensing

0 Small size: 1.8' tall x 1.9' wide x 0.6'' thick and 0.9 oz.

External thermistor probe on a ﬂexible cable

(ordered seperately)

Precision thermistor and converters require no calibration

Standard cable lengths are 1, 2 and 6 feet

Optional submersible case rated to 400' depth

Nonvolatile EEPROM memory retains data even when the battery has

been removed

Stores up to 1800 measurements

O 42 preselected intervals from 0.5 second to 4.8 hours, corresponding to
deployment durations up to 360 days

0 Start the logger, readout and plot the data with BoaCar‘ or LogBook‘ software
for Windows or Mac

0 Blinking light conﬁrms operation

0 Data readout in less than 30 seconds

0 Dataexpm'tabletospreadsheetprogramsarotuaExcelmtc.)

Overview

TheHOBO’X'l‘temperatureloggeruﬁlisesanexternalM
mister cable which is ideal for recording temperature in hard-to-
reachlocations. BoxCar‘orLogBook’sottwarecanstart,read-
outandgraphdataforWindowsor-Mac.

 

 

   

 

Temperature accuracy

T'heHOBOXT’smaximumerrorisshowninPlotA. T‘hiserror
assumes that all contributing factors to the error are at their
maximum values and are aligned so their values add together.
These errors include thermistor error, resistor value errors, and
quantization error (step errors in the digital representation of
thetemperature). In atypical loggerallerrorswillbesubstan-
tially lower.

Temperature resolution

The HOBO XT‘s resolution (diﬁ‘erence between the temperature
steps) is shown in Plot B. Note the similarity between the
resolution (Plot B) and accuracy (Plot A) for the widest range
loggers, where most of the error is due to the quantization error.

Thermal time constant

The thermal time constant (90% response to a step change in
temperature) of a HOBO XT's external sensor shows a time
constant of less than three minutes in air and less than ﬁfteen

summit: in stirred water

@1133 T°'= (508)563r9000 o momma-9477 . 886: (508) mm 0 email: salesOonsetcompeom
mm 536MocArthurBlvd. e aoxaaso e Pocasset.MA0255¢-3450

Figure A-2. Speciﬁcation of the Hobo® temperature data logger manufactured by Onset
Inc.

103

APPENDIX

Relative humidity logger

StowAway‘“ RH

Features
Rated 5% to 95% RH, non-condensing
Accuracy 156% tolerance at room temperature
Two year battery life (user replaceable)
N onvolatile EEPROM memory retains data even when battery
is removed
Memory conﬁgurations are 2K, SR or 32K storing 1800, 7944 or
32,520 measurements
Safe operating temperature range of electronics is 0°C to 60°C,
non-condensing
Small size: 1.8' tall x 1.9“ wide x 0.6' thick and 0.9 oz.
Start the logger, readout and plot the data with LogBook' software
for Windows or Mac
42 preselected intervals from 0. 5 second to 4. 8 hours,
corresponding to deployment durations up to 2 years
Blinking light conﬁrms opera
Alarm indication
Programmable delayed start (up to three months)
Push button triggered start
Multiple sampling with minimum, maximum or averaging
Data exportable to spreadsheet programs (Lotus, Excel, etc.)
DOS and batch utilities available

Overview
The StowAway" RH is a general purpose, durable, and reusable relative humidity logger. LogBook‘
software for PCs or Macs makes launching, readout, plotting, and analysis a snap.

Calibration
Each StowAway RH is individually tested at ten relative humiditics ranging from less than 10% to greater than 90%.
Its calibration is permanently stored in the StowAwsy RH. ensuring accurate measurements.

 

Temperature dependence
The StowAwale-l sensorhss sternpcnmre dependence oi'0.22% RHpcrdegreeC. Alcadegrec Cexcursion from
room (carpenters will add about 2% RH error.

Sensor Speciﬁcations

Response time: 2 minutes

Repeatability: less than 2% (cum tarmac)
Storage temperature: 40°C to +60°C
Temperature coefﬁcient: 0.22% per °C

@1159: Tel:(508)563—9&1'J 9 Fax (508)563-9477 0 ass (503)5os2259 e _mi: salosoomercomp_.__ com
WW 536MocArthu18h/d. o 30x3450 e Pocosset.MA02559-3450

Figure A-3. Speciﬁcation of the StowAwayTM relative humidity data logger
manufactured by Onset Inc.

APPENDD(

Relative humidity logger
HOBO® RH

Features

0 Rated 6% to 95% RH, non-condensing

- Accuracy 15% tolerance at room temperature

The year battery life (user-replaceable)

Safe operating temperature range of logger is 0°C to 60°C
Small size: 1.8“ tall x 1.9" wide x 0.6" thick

and 0.9 oz.

Nonvolatile EEPROM memory retains data even when

the battery has been removed

Stores up to 1800 measurements

Start the logger, readout and plot the data with BoxCar" or
LogBook‘ soitware for Windows or Mac

0 42 preselected intervals from 0.5 second to 4.8 hours,
corresponding to deployment durations up to 360 days
Blinking light conﬁrms operation

Data readout in less than 30 seconds

0 Data exportable to spreadsheet programs (Lotus, Excel, etc.)

 

 

Overview

The HOBO. RH is a general purpose, relative humidity logger that is both du-
rable and reusable. Its sensor resists chem: cal corrosion by chlorine, acetone,
pentane, xylene, formaldehyde, ammonia, hospital germicides and freon. The
HOBO RH has an operating range of 0°C to 60°C and rs designed for a non-
condensing environment.

Calibration

Each HOBO RH is individually tested at ten relative humidiﬁes ranging ﬁ‘om
less than 10% to greater than 90%. Its calibrations are permanently stored in
the HOBO RH ensuring accurate measurements.

Temperature dependence
The HOBO RH's sensor has a temperature dependence of 0.22% RH per degree
C. A ten degree C excursion from room temperature will add about 2% RH error.

Sensor Specifications

nse time: 2 minutes
Repeatability: less than 2% (constant mpg-stare)
Storage temperature: -40°C to +60°C
Temperature coefficient: 0.22% per °C

@nggt Tel: (503) 563-9000 9 Fax: (508) 563-9477 9 ass: (503) 553.2259 0 email: salesOonsetcompcom
536MocAr‘lhurBNd. e Box3450 e Pocasset.MA02559-3450

 

Figure A-4. Speciﬁcation of the Hobo® relative humidity data logger manufactured by
Onset Inc.

105

APPENDIX

Light intensity logger
StowAway‘” Ll

Features

0 Calibration ranges from less than 0.001 lumen/square foot to
over 1000 lumen/square foot

Wide spectral response and wide dynamic range

The year battery life (user replaceable)

Nonvolatile EEPROM memory retains data even when battery
is removed

0 Memory conﬁgurations are 2K, SR or 32K storing 1800, 7944
or 32,520 measurements

Safe operating temperature range of logger' is 40°C to +75°C,
non-condensing

Small size: 1.8" wide x 1.9” tall x 0.6” thick and 0.91 or.

Start the logger, readout and plot the data with LogBook'
soﬁware for Windows or Mac

42 preselected intervals from 0 5 second to 4. 8 hours,
corresponding to deployment“ durations up to 2 years

 

- Blinking light conﬁrms operati

0 Alarm indication

- Programmable delayed start (up to three months)

- Push button tri start

0 Multiple sampling with minimum, maximum ore veraging

0 Data exportable to spreadsheet programs (Lotus, Excel, etc.)

0 DOS and batch utilities available

Overview

The StowAway" LI is designed as a durable, reusable. inexpensive, general purpose light intensity
logger. Sensitivity ranges from 0. 001lum umens/square foot ootto 10001umenslsquare foot. LLogBook‘ sottware
for PCs or Macs makes launching, readout, plotting and analysis a snap.

Temperature dependence

'IheStowAwayIJiscalibratedatroomtempei-atm'e. 'I‘heleggerwillreadhighfortamperaturesabovemom
temperattn‘eandlowt‘ortemperatm'esbelow. Themorisapproximatelyatactoroftwoforeveiy25°c
change. ThismeansitwinmadatactmoftwohighatO'Candafactoroftwolowatwc.

Calibration

The StowAway LI is roughly calibrated for incandescent sources. N1 sunlight is about 10,0001umens/
square foot, ofﬁce lighting is about 50 lumens/square foot, and full moonlight is about 0.03 lumens/
square foot. The StowAway LI‘s range goes from less than 0.0011umens/square foot to about 1,000 lu-

mens/square foot.

Spectral re

The sensitivity of the StowAway LI’s photo sensor extends into the near

i as shown in Plot A. Althou thisis useﬁil' in many applications, it

also means that the logger' s responsivity is strongly dependantn on
spectral characteristics of the light it is measuring. For example, the
logger will read about a factor ten low when reading ﬂuorescent lighting

Angular dependence
The angular dependence of the logger from 0° to 45" vertical resembles the
curve for cosine. After 45" the curve drops of? more rapi y.

 

@nsgt Tel: (mimomo 9 Fax: (508) 563-9477 9 ass: (508)563-2269 9 email: salesOonsetcormcom
mun-r 536MocArthurBlvd. . Box 3450 o Pocasset.MA02$59~3450

Figure A-S. Speciﬁcation of the StowAwayTM light data logger manufactured by Onset
Inc.

106

APPENDD(

Table A-1. Average, maximum, and minimum daily temperature and relative humidity
of chambers and controls and climatogical data of the near by NOAA station from snow
melt until August 18th for the years 1995 and 1996.

 

 

 

1995
NOAA Control OTC

pm Te—mperatu'r—e W rm

(0"!) (° C) Humidity (° C) Humidity (° C) Humldﬁv
Day meanwrrmmeanmnmnmnmmmeanm mmnnoxminmeanmx min
7qu 0 4.8123 2.7 88.3 100— 4.0 7.1 1.3-- — — 6.0 12.3 21"- - -
8-Jul 1 5.8171 2.0 74.7 95.8— 9.7 20.1 1.3-- - -- 10.2 212 1.8- -- --
9401 0 9017.1 3166.9 100— 6.8134 3.0-- - —- 8.8 202 3.4-- —- —
10-Jul O 3.9 8.0 2.7 82.2 97.9— 4.6 7.8 2.1 -- -- .- 7.1 13.9 25"- -- -
11qu 0 2.7 5.0 1.8 81.2 90.4— 3.1 8.8 1.0-“ — - 5.6 12.5 1.8"' -- —-
12-Jul 0 2.0 3.2 1.5 86.5 91.0— 2.4 5.1 0.3- —— —. 4.9 11.3 12 -- - -
13-Jul O 1.8 4.8 -o.187.6 100— 4.4 8.8 -o.8-- -- — 9.2 18.7 -0.1-- — --
14qu O 4.1 8.0 1.7 79.9 100— 5.5 9.8 0.3“- -- - 9.2 18.8 1.8"' -- -
15-Jul 1 6.2109 4.4 87.2 99.8— 7.8133 3.4-- — — 10.8 19.1 3.9-" - --
16-Jul 0 4.7 3.1 1.8 86.6 100— 6.0105 1.6""" -- -- 9.3 19.2 2.7- -- --
17-Jul 3 8.1 10.9 5.3 93.7 100— 7.6102 4.3-“ — -— 8.2 10.8 4.8"' -- -
18-Jul 3 8.9190 4.0 89.1 100-- 9217.6 2.7- -- .- 9.7 19.1 3.3 -- -— --
19-Jul 6 3.4104 0190.1 100— 2.7 8.8 -02-- - -- 3.4 8.4 0.0“- - --
20qu 0 2.9 5.1 1.5 80.2 91.0 -- 2.8 5.4 0.8 -- — —- 3.6 7.5 0.9 -- -- --
21qu O 2.0 3.8 0.9 85.9 100— 2.3 5.3 02- -- -— 4.6 11.5 0.4-- - -
22-Jul O 1.6 3.2 0.3 91.8 100— 2.2 4.9 .02"- - — 4.1 9.7 0.0- -- -—
23-Jul 0 2.4 4.1 1.3 82.0 95.1 —- 2.9 72 -o.3-- - -. 5.3 12.3 0.8- - --
24-001 0 0.9 32 -o.188.1 100— 1.2 42 -1.0-- — —- 3.3 8.4 0.3-- -- --
25-Jul 0 0.0 1.7 .1.0 91.8 99.1 - 0.7 3.9 -1.7-- - -— 3.4 10.0 -1.4-- - -
26-Jul O -0.4 0.7 -1.8 94.2 99.8— 0.6 3.5 -2.3-- - -- 2.8 9.1 -1.8-- - -
27-Jul O 0.9 1.8 02 92.0 99.3— 1.4 4.8 -o.8-- -— -- 5.0 14.8 0.3— -—- -
28-Jul 0 0.8 2.4 -o.3 91.3 99.8— 2.3 5.5 -12-- — ..- 5.3 12.0 -12-- -- -
29qu 0 1.4 2.4 0.4 90.2 97.8 - 1.7 5.0 -o.3 -- — .— 3.5 9.1 -o.2 -- -- --
30qu 2 1.4 2.3 0.7 90.6 95.7- 1.2 2.3 0.1-- — .— 1.7 3.7 0.1"- -- -
31-321 1 2.5 3.8 1.7 90.7 100— 3.0 5.7 1.8-— -- -- 4.1 9.5 1.8- -- --

-—datanotavallable

107

 

AHHHEDUDEX

Table A-1 (cont’d).

 

 

 

 

1995(conﬁnued)
NOAA Control OTC

Pm Te‘mperatur'i HGIBWG mature Relatlv'e—

(cm) (°C) Humidity (° C) Humidity (° C) Humidity
Day meanwminmean maxmlnmeanmaxminmean max minmean maxminmean max min
1-Aug O 4.2 8.5 3.3 92.1 100— 4.2 8.8 2.4-- — — 5.6 14.3 2.6’“ — --
2-Aug O 2.7 3.8 2.0 92.7 100— 2.1 4.8 -o.8-- -. -- 3.8 10.1 -02-- -- --
3-Aug 0 1.1 2.4 -o.2 92.1 100— 1.7 5.4 -0.9-- — -- 3.5 11.7 -0.8-- -- --
4-Aug 0 0.9 2.2 -o.5 96.0 100— 1.1 4.8 -1.8-- -- -- 3.2 10.2 -1.5-- -- --
5-Aug O 1.0 2.4 0.0 93.0 100... 1.6 4.7 -1.1-- — .- 3.2 10.5 -o.9-- —- —-
6-Aug 0 2.3 3.4 1.4 93.0 100— 2.4 4.9 0.5- .- .— 4.1 92 0.5- -- -
7-Aug 0 1.3 3.8 0.0 98.3 100— 1.3 3.9 -o.7-- - .- 3.5 9.1 -0.8-- -— -
8-Aug O 1.8 3.8 0.5 93.4 100- 3.2 8.5 -0.3-- .- -- 6.1 12.5 -0.4-- -- --
9-Aug 0 2.7 3.8 1194.7 100— 3.0 8.8 0.4-- .- .— 4.9 11.9 0.5- - --
10-Aug O 3.3 5.1 1.3 96.5 100— 3.7 8.5 0.5— ~— -- 5.4 10.0 O.6""' -- -
11-Aug O 3.4 4.9 2.7 97.9 100- 3.0 5.3 0.9- -- - 4.0 7.4 1.3-- —- --
121109 0 2.6 4.3 1.5 96.9 100— 3.3 8.8 0.9-- -- .— 4.9 122 1.0 -- -- --
13—Aug 0 2.6 4.0 1.5 96.7 100- 2.6 5.1 0.8"" -- -— 3.9 9.0 0.9“- - -
14-Aug O 2.2 5.5 0193.5 100— 2.9 8.0 -o.8-- —- - 3.9 11.5 -0.8-- — -
15-Aug 0 3.0 8.1 -0.3 85.0 100— 4.6115 -1.0-- - -- 5.7 13.8 -0.7-- .- --
18-Aug 0 7.7 9.8 5.8 83.5 95.9- 6.8118 42-- -- .- 7.3 13.5 3.9-- -- --
17-Aug 0 6.2140 22 88.4 100— 6.6170 0.1- -- .- 8.2 18.5 0.1-- —- --
18-Aug 1 7.01o.0 4.8 97.0 100— 5.3100 0.0-- .- —. 5.6 10.2 0.5-- —- -
19-Aug 1.0 4.9 -05 91.6 100— 0.1 32 -1.9-- — .— 1.0 8.2 2.0- - --
20-Aug 0.9 3.8 -o.7 89.5 99.8— 1.2 5.0 -1.9-- - -— 1.6 8.8 -1.9-- -- -—
214199 2.0 4.8 3392.0 100— 2.5 7.7 -5.8-- -- - 3.8 10.9 -1.8-- -- —-

~datanotavallable

108

APPENDIX

Table A-1 (cont’d).

 

 

 

 

 

1996
NOAA Control OTC
Pap WpemMrB— Below W HeTiﬁ've TeWperature R8156?
(cm) (°C) Humidity (°C) Humidity (°C) Humidity
Day memwninmnmmwnmxnmmeanmxnmmeanmaxmmmeanmxm
10-Jun 0 -1.1 1.2 -3.1-- - - 0.3 2.5 -2.7-- - -. 3.3 8.4 -22 71.7 79.2 81.3
11-Jun O -0.9 1.4 -3.2-- — -- -0.2 2.5 -2.9-- .- ..- 0.9 4.8 -2.7 77.4 85.9 88.8
12-Jun O 0.6 3.0 85 -- — — 1.0 3.8 -0.8 -- -- ..- 2.2 7.8 -0.5 73.7 87.5 80.5
13—Jun O 1.2 3.4 0.4 97.9 100— 2.3 4.9 0.3- —- —. 4.5 11.7 0.3 67.8 81.8 57.9
14-dun O 1.5 3.8 0.0 94.7 100— 2.6 5.8 -02-- -- - 5.5 12.4 0.7 64.6 79.9 58.7
15-Jun O 1.9 5.4 -02 91.9 97.8— 4.1 7.7 02-- -- - 6.8 13.8 0.3 64.0 79.0 52.5
16-Jun 0 3.8132 0.7 89.1 97.2- 6.5180 0.4 84.2 88.8 80.4 8.7 18.9 0.9 75.6 83.8 88.4
noun 0 5.4 8.0 2.5 88.2 96.6- 6.1 102 2.5 78.0 88.3 83.7 6.6 12.8 2.9 66.7 88.0 47.5
188m 0 5.9118 3.0 85.0 99.3— 9.4185 3.8 62.7 82.8 37411.5 20.9 3.7 50.8 72.3 33.1
19-Jun 012.4 22.7 4.5 67.4 92— 14.4 23.7 3.7 76.5 90.3 83116.2 24.9 8163.6 85.4 48.3
20-Jun 3 8.4 22.8 2.7 76.5 100— 9.3188 3286.9 92.7 78811.0 20.4 3.8 73.9 87.8 80.8
21-Jun O 5.5 9.0 3289.7 100— 5.7101 1.1891 94.0 83.1 7.1 18.4 1.3 74.8 87.1 88.8
22-Jun O 0.9 5.4 -o.9 95.3 1oo-- 3.3113 -1.3 85.8 93.5 73.1 5.1 11.5 -1.177.2 89.7 88.4
23-Jun 0 5.3108 0.6 85.7 100— 4.8 9.8 0188.2 94.9 78.7 5.1 9.5 -o.3 76.0 89.7 82.0
24-Jun 1 0.5 2.4 -0.6 88.2 97.3— 1.8 5.7 0.8 89.3 94.0 83.1 2.7 8.0 -1.5 74.6 88.7 58.4
25-Jun O -0.1 2.0 -1.5 91.9 99.8— 1.3 4.9 -1.5 89.6 94.9 82.5 3.7 11.0 -1.5 77.6 87.8 87.2
26-Jun 0 0 1.4 4.1919 99.1 -- 1.0 3.7 -0.8 91.4 95.4 882 2.2 8.7 -0.8 77.3 89.2 83.2
27-Jun 0 ~02 12 -0.8 95.0 99.5- 0.9 3.8 -0.9 88.2 93.5 79.8 3.0 8.4 -0.8 68.6 84.3 48.0
28-Jun 0 ~09 0.3 -22 94.0 99.5— 1.1 5.0 -1.0 82.4 91.1 88.7 5.2 14.7 -02 64.5 82.7 41.9
29-Jun O -1.4 0.8 -2.7 89.9 98.2 —- 1.6 8.9 -2.5 76.2 94.9 82.5 5.4 18.4 -1.6 61.3 88.0 402
30-Jun 0 3.6 9.2 0277.1 97.9— 7.1 13.4 -1.3 89.6 94.9 80410.0 21.5 -o.3 71.1 88.5 45.2
1qu 0 2.3105 -1.0 87.3 100— 2.5 7.1 8.8 76.0 93.5 58.8 5.7 17.0 -o.3 62.2 87.0 45.8
2-Jul 0 7.7193 0.5 81.4 100— 11.3 21.4 2.7 85.0 89.8 71012.6 242 1.3 69.2 79.9 50.4
3qu O 5.6 8.4 1.8 85.4 972—- 7.2118 3190.5 94.4 85.7 9.4 20.8 2.5 77.6 87.0 88.4
4-Jui O 5.2 7.2 32 92.3 972— 5.2 8.8 3.8 82.9 90.7 742 6.4 10.9 2.8 68.5 84.9 51.4
5-Jul 0 3.7 5.0 2.8 86.6 93.1 - 5.1 9.0 2189.8 982 85.7 7.2 15.8 1.5 74.1 87.8 83.0
6-Jul 2 2.6 4.7 1292.3 100— 4.2 7.7 1.2 92.8 95.7 87.7 5.5 12.8 -02 80.0 88.1 88.1
7qu 0 3.4 8.0 2.0 95.8 100— 4.4 8.3 2.9 93.7 97.5 89.7 5.6 13.4 1.4 82.2 90.3 73.1
8-Jul 1 3.3 7.8 1.8 94.7 100— 3.9 7.8 0.8 89.7 97.5 80.9 4.1 7.9 -o.2 72.5 90.8 54.4
9-Jul 0 0.6 3.0 -0.8 94.5 100— 2.0 5.2 -o.5 88.5 94.0 80.4 3.4 10.8 -1.3 70.9 84.9 54.9
104101 0 -0.3 1.4 -1.4 89.6 95.9— 1.3 5.3 -1.1 89.7 94.4 83.8 3.8 11.4 -0.9 73.7 87.0 83.7
11-Jul 0 -O.4 1.9 -1.5 93.9 99.7— 1.4 5.0 -1.2 90.1 98.2 75.3 3.6 9.2 -1.0 73.9 91.8 52.0
12-Jul 0 0.8 4.1 -1.2 95.2 100— 4.6130 -1.0 83.1 91.1 72.5 6.6 18.7 -1.3 67.1 83.8 47.5
13-Jul 8 913.7 3.7 83.3 93.8— 12.0173 5.5 91.1 98.2 85713.7 24.8 4.0 74.4 88.1 57.9
14qu 0 8.4133 4293.0 100-- 9.8 15.0 3.8 91.1 98.8 83812.3 24.0 4278.3 91.8 88.7
15—Jul 4 5.2115 1.4 91.9 100— 4.2 8.0 1.9 96.7 98.2 93.0 5.4 14.1 2.0 90.6 94.8 89.2
16-Jul 3 5.8 8.8 1.5 94.6 99.3— 6.8 8.3 4.8 97.0 99.0 89.7 6.7 8.0 4.8 90.9 95.3 81.0
17qu 1 5.5 6.7 4.8 99.1 100— 6.6103 4.4 92.6 98.2 83.1 6.6 10.4 4.3 84.2 93.8 73.7
18-Jul 1 7.4107 1.7 88.2 99.3— 5.9121 0.0 96.6 99.4 89.1 6.5 12.1 1.3 88.9 94.3 81.0
-- data not available

109

APPENDIX

Table A-1 (cont’d).

 

 

 

 

1996 (continued)
NOAA Control OTC
Pm Wm Helaﬁ’ve Temperature Tieiaﬁﬁ'
(cm) (°C) Humidity (°C) Humidity (°C) Humidity
Day meanmaxminmeanmaxminmeanmaxminmeanmax mianaxminmeanmax min
19-Jul 2 1.8 9.9 -12 98.4 100— 5.8143 -1.4 96.7 99.0 91.8 6.2 14.2 8.9 90.4 95.3 87.1
20-Jul 0 6.9142 0.9 93.1 100— 4.6103 0194.3 98.5 88.1 4.8 10.1 0580.9 94.3 85.0
21qu O 0.9 6.2 .1.1 95.6 100— 1.8 8.1 -0.8 94.7 98.5 88.1 4.1 10.8 -o.4 85.8 94.8 72.5
22qu 0 3.8 8.7 0.0 95.4 100 - 4.5 8.3 0.2 87.9 94.4 79.3 5.0 9.4 0.3 72.4 90.2 53.7
23qu O 0 3.8 -1.6 85.1 95.8— 2.0 5.8 -1.9 82.0 95.8 84.3 5.0 14.8 -2.0 74.1 89.7 58.4
24301 0 7.4198 1.0 83.9 94.8— 12.7 20.8 2.7 81.0 94.8 83112.6 21.1 2.8 72.9 90.3 54.3
25qu 0155192102611 90.4—- 13.0 20.2 7.5 88.1 98.1 78513.1 20.4 7.4 81.8 93.3 70.7
26-Jul 010.7170 7188.2 100— 12.8172 7.8 85.5 93.9 77012.4 17.1 7.8 76.0 89.2 83.8
27qu 012.5173 6.2 78.4 94— 9.9147 3.8 86.7 93.4 79710.5 15.9 3.5 78.5 88.1 702
28~Jul O 4.3 7.4 2.7 81.8 99.3— 4.2 6.7 22 82.3 92.0 89.8 4.4 7.9 2.0 71.5 89.2 53.2
29-301 0 3.8 8.0 2277.7 93.8— 5.4 9.9 1.8 82.2 94.8 70.8 6.9 14.3 1.8 62.3 90.8 42.9
30qu 0 3.9 6.6 1.7 82.7 95.8— 4.5 9.4 4.4 86.7 95.3 77.0 9.2 19.2 -3.6 73.0 88.1 51.4
31qu O 3 5.9 0.0 82.4 95.1— 3.9 9.2 3186.2 98.1 77.5 5.9 16.0 -1.9 71.4 88.5 58.1
1-Aug 0 2.7 7.2 0284.7 97.2- 4.7100 -1.0 96.8 99.1 94.8 6.4 14.8 0.8 87.1 91.8 85.4
2-Aug 0 2.8 7.9 0.4 92.4 100-- 2.3 4.8 0.5 94.9 99.1 88.8 2.7 8.2 0.5 81.9 93.3 82.8
34109 0 4.6102 2.3 97.1 100— 7.1 142 2.3 94.1 99.1 87.8 8.3 19.4 2284.5 93.8 72.5
4-Aug O 6.1 9.2 3.5 93.3 100— 6.0103 3.5 92.6 98.1 87.1 6.3 11.5 3.4 79.4 93.3 832
5-Aug 0 4.3 7.2 2.8 89.7 97.9-— 4.2 7.4 0.9 93.6 98.1 89.0 5.9 12.4 0.9 80.5 93.3 84.9
6-Aug O 2.6 4.3 1.0 91.5 99.3— 2.8 7.3 -2.9 93.3 98.5 85.8 4.4 12.3 -2.5 84.0 93.3 75.4
7-Aug 0 2.1 4.8 0191.0 100— 3.1 5.8 8.8 94.7 98.1 88.8 3.5 7.0 8.8 86.3 93.3 74.8
6-Aug O 3.7 6.0 2190.9 99.3.- 4.5 8.7 2297.1 99.5 93.8 4.7 9.7 2287.8 93.8 82.2
9.409 0 4.4 7.0 1.7 94.9 100— 3.1 4.8 0.5 91.2 95.1 85.8 3.7 6.0 0.5 81.7 90.8 71.8
10-Aug 0 0.5 2.4 -o.9 86.1 95.1— 0.8 3.1 -0.8 91.8 95.8 89.0 1.5 5.7 -0.9 82.9 90.8 75.4
11-Aug O 0.2 2.0 -1.2 85.2 95.3— 0.9 4.2 -1.5 90.4 94.8 81.9 1.4 8.0 -1.8 76.5 90.2 53.8
12-Aug O 3 8.5 0.7 87.7 95.6— 5.3104 1.8 96.2 99.1 91.0 6.8 17.7 1.8 87.6 94.8 81.0
13-Aug 3 4.6 7.4 2192.5 99.3— 4.8 8.0 1297.1 99.5 90.4 5.0 8.8 1.1909 95.3 88.0
14-Aug 4 6.8112 4092.0 98.8-— 6.8110 1,091.1 98.7 79.7 6.7 11.0 1.081.7 95.3 88.1
15-Aug 0 2.4108 8.2 90.1 100— 1.8 8.1 -o.9 92.1 98.3 84.1 3.3 10.2 -1.179.3 91.8 83.8
18-Aug 0 0.5 3.0 -1.o 86.8 98.5— 2.3 7.3 -1.0 96.1 99.5 91.0 3.7 11.9 -12 86.5 96.2 79.3
17-Aug 0 2.5 8.0 8.8 92.6 100— 2.0 8.1 2191.6 98.7 80.8 2.6 8.1 -2.4 78.0 94.3 57.9
16-Aug 0 1.1 4.1 0292.3 999-— 3.3 8.8 0.0 98.1 99.5 94.8 5.5 14.7 0.0 89.1 95.3 84.4
--data not available

110

APPENDIX

Table A-2. Average active layer thickness and standard deviation for the ﬁeld seasons
1995 and 1996 (N for OTC = 2.4; N for control = 96).

 

 

ActiveLayerThickness
Day 1995 1996
OTC Control OTC Control
mean Std mean std mean Std mean 810
10-Jun .- --- -- 3.0 0.54 1.5 023
13-Jun ------ 4.1 0.58 3.0 0.29
14-Jun -- 5.5 0.58 4.3 0.25
15-Jun ------ 7.0 0.88 5.6 0.24
16-Jun 8.8 0.85 7.6 0.26
17-Jun 10.4 0.71 8.8 0.27
18-Jun 12.0 0.75 10.9 0.30
19—Jun 15.3 0.94 14.1 0.38
20—Jun 17.1 0.93 16.1 0.39
21-Jun 18.5 0.91 17.6 0.41
23-Jun -- 20.3 0.87 19.3 0.50
24-Jun --- -- --- -- 23.0 1.01 21.6 0.48
25-Jun 2.8 0.25 2.2 0.13 --- -- --- -.-
26-Jun 2.4 0.24 2.1 0.13 --- ---
27-Jun 2.3022 2.0 0.25 -- -.-
28-Jun 2.5022 1.9 0.13 --- m
29-Jun 3.6023 2.9 0.13 --- ---
30-Jun 4.00.23 3.5 0.15 --- ... --- ..-
1-Jul 5.70.22 5.0 0.18 25.2 0.62 23.9 0.51
2qu 6.80.24 6.2 0.17 .. ..-
3-Jul 8.60.34 7.8 0.21 ---
4-Jul 9.70.38 9.3 0.22 --- ---
5-Ju|10.5 0.38 9.9 0.25
6-Jul 11.5 0.47 11.4 0.27 --- --- ---
7-Jul 13.50.85 12.6 0.29
8-Jul 14.00.66 13.9 0.31 --- ---
9qu 16.2 0.73 16.4 0.33 ..
10-Jul --- -- --- -- 29.9 0.92 29.2 0.61
16-Jul 27.0 0.98 25.0 0.50 --- -- --- -.-
20-Jul --- -- -- 36.2 1.08 36.5 0.64
23-Jul 33.31.07 30.7 0.59 -- --- --
30-Jul 32.41.23 29.8 0.60 44.4 1.93 42.7 0.66
--- no data available

111

APPENDIX

Table A-2 (cont’d).

 

Active Layer Thickness
Day 1995 1996
OTC Control OTC Control

meanStdrmeanstd 111880810 111880510

6-Aug 31.81.13 29.0 0.55 --- .. --- ..

94109 -- -- 45.8 1.27 45.6 0.95

13-Aug 35.31.31 33.7 0.67 ... ..

20-Aug 38.01.42 35.8 0.74 «- ---

21'AU9 -- -- 50.5 1.54 48.1 0.90

26-Aul36.81.13 35.9 0.72 --- .. --- ..-
-- no data available

112

 

 

LITERATURE CITED

113

LITERATURE CITED

Anderson, .l.M. (1991) The effects of climate change on decomposition processes in
grassland and coniferous forests. Ecological Applications. 1:326-47.

Anisimov, O.A., N.I. Shiklomanov, and RE. Nelson. (1997) Global warming and active-
layer thickness: results from transient general circulation models. Global and
Planetary Change. 15 :61-77.

ARCU S. (1993) Arctic System Science: A Plan for Integration. Fairbanks, Alaska.
September 1993. 60 pp.

Arrhenius, S. (1896) On the inﬂuence of the carbonic acid in the air upon the
temperature of ground. Philosophical Magazine. 41:237-75.

Babb, T.A., and D.W.A. Whitfield. (1977) Mineral nutrient cycling and limitation of
plant growth in the Truelovc Lowland ecosystem. In Bliss, L.C. (ed.) Truelove
Lowland, Devon Island, Canada: A High Arctic Ecosystem. University Alberta
Press. Edmonton, Canada. p. 589-606.

Bartlein, PJ., C. Whitlock, and S.L. Shafer. (1997) Future climate in the Yellowstone
National Park region and its potential impact on vegetation. Conservation
Biology. 11:782-92.

Bay, C. (1995) International Tundra Experiment, Barrow, Alaska: BPRC Technical
Report 95-02. Byrd Polar Research Center, Ohio State University. Columbus,
Oh. 51 pp.

Bay, C. (1996) International Tundra Experiment, Barrow, Alaska: Plant and Physical
Response Under Open-Top Chambers on Dry Tundra. BPRC Technical Report
96-03. Byrd Polar Research Center, Ohio State University. Columbus, Oh. 211

PP-

Bazzaz, F.A. (1990) The response of natural ecosystems to the rising global C02 levels.
Annual Review of Ecology and Systematics 21: 167-96.

Billings, W.D. (1974) Arctic and alpine vegetation: plant adaptations to cold summer
climates. In Ives, J.D., and R.G. Parry. (eds.) Arctic and Alpine Environments.
Harper & Row. Methuen, London. p. 403-43.

Billings, W.D. (1992) Phytogeographic and evolutionary potential of the arctic ﬂora and
vegetation in a changing climate. In Chapin, F.S., R.L. Jefferies, J.F. Reynolds,
G.R. Shaver, and J. Svoboda. (eds.) Arctic Ecosystems in a Changing Climate:
An Ecophysiological Perspective. Academic Press Inc. San Diego, Ca. p. 91-109.

114

Billings, W.D. (1995) What we need to know: some priorities for research on biotic .
feedbacks in a changing biosphere. In Woodwell, G.M. MacKenzie, and T. Fred.
(eds.) Biotic Feedbacks in the Global Climatic System. Oxford University Press.
New York, NY. p. 378-92.

Billings, W.D. (1997) Arctic phytogeography. In Crawford, R.M.M. (ed.) Disturbance
and Recovery in Arctic Lands: An Ecological Perspective. Kluwer Academic
Publishers, Boston, Massachusetts. p. 25-45.

Bliss, LC. (1962) Adaptations of arctic and alpine plants to environmental conditions.
Arctic. 15:117-44.

Bliss, LC. (1988) Arctic tundra and polar desert biome. In Barbour, M.G., and W.D.
Billings (eds.) North American Terrestrial Vegetation. Cambridge University
Press. Cambridge. p. 1-32.

Bliss, L.C., G.M. Courtin, D.L. Pattie, R.R. Riewe, D.W.A. Whitfield, and P. Widden.
(1973) Arctic tundra ecosystems. Annual Review of Ecology and Systematics
4:359-99.

Bonan, G.B., H.S. Herman, and D.L. Urban. (1990) The sensitivity of some high-latitude
boreal forests to climatic parameters. Climatic Change. 16:9-29.

Box, 15.0. (1981) Macroclimate and Plant Forms: An Introduction to Predicitve
Modeling in Phytogeography. The Hague. Boston, Ma. 258 pp.

Britton, Max E. (1957) Vegetation of the Arctic Tundra. In Hansen, H.P. (ed.) Arctic
Biology. Oregon State University Press. Corvallis, Or. p. 26-61.

Brown, J. (1997) Frozen Ground. The News Bulletin of the International Perfafrost
Association. Cold Regions Research And Engineering Laboratory. Hanover, NH.
21:1-36.

Brown, J., and J.T. Andrews. (1982) Environmental and societal consequences of a
possible C02 - induced climate change. Carbon Dioxide Effects Research and
Assessment Program. U.S. Department of Commerce. Springﬁeld, Va. 30 pp.

Brown, J ., K.R. Everett, PJ. Webber, S.F. MacLean, and DP. Murray. (1980) The
coastal tundra at Barrow. In Brown, J ., P.C. Miller, L.L. Tieszen, and FL.
Bunnell (eds.) An Arctic Ecosystem: The Coastal Tundra at Barrow, Alaska.
Dowden, Hutchinson, & Ross inc. Stroudsburg, Pa. 12:1-29.

Brown, J., and RV. Sellmann. (1973) Permafrost and coastal plain history of arctic
Alaska. In Britton, M.E. (ed.) Alaskan Arctic Tundra, Arctic Institute of North
America Technical Paper No. 25, Proceedings of the 25'” Anniversary
Celebration of the Naval Arctic Research Laboratory. Arctic Institute of North
America. Washington, DC p. 31-47.

115

Bryant, LP. and PB. Reichardt. (1992) Controls over secondary metabolite production
by Arctic woody plants. In Chapin, F.S., R.L. Jefferies, J.F. Reynolds, G.R.
Shaver, and J Svoboda. (eds.) Arctic Ecosystems in a Changing Climate: An
Ecophysiological Perspective. Academic Press, Inc. San Diego, Ca. p. 377-386.

Budyko, M.I., A.B. Ronov, and AL. Yanshin. (1987) History of the Earth 's Atmosphere.
Springer-Veﬂag. Berlin. 139 pp.

Callaghan, T.V., B.A. Carlsson, and NJ .C. Tyler. (1989) Historical records of climate-
related growth in Cassiope tetragona from the arctic. Journal of Ecology.
77 :823-37.

Callaghan, T.V., U. Molau, W.C. Oechel, T. Gilmanov, B. Maxwell, MJ. Tyson, B.
Sveinbojorrnsson, and J .I. Holten. (1995) Global Change and Arctic Terrestrial
Ecosystems. Springer-Vedag. New York, NY. 329 pp.

Cane, M.A.,A.C. Clement, A.Kaplan, Y.Kushnir, D.Pozdnyakov, R.Seagen, S. Zebiak,
R.Murtugudde. (1997) Twentieth-century sea surface temperature trends.
Science 275:957—60.

Cantlon, 1.13. (1961) Plant Cover in Relation to Macro-, Mesa-, and Micro- Relieﬂ Final
report to the Arctic Institute of North America on Contracts ONR-208 and ONR-
212. 128 pp.

Catovsky, S. (1998) Functional groups: clarifying our use of the term. Bulletin of the
Ecological Society of America. 79: 126-27.

Chapin, F.S. (1984) The impact of increased air temperature on tundra plant
communities. In McBeath, J.H., G.P. Juday, G. Weller, and M. Murray. (eds.)
The Potential Eﬂ’ects of Carbon Dioxide-Induced Climatic Changes in Alaska.
University of Alaska. Fairbanks, Ak. p. 143-148.

Chapin, F.S. (1987) Environmental controls over growth of tundra plants. Ecological
Bulletins. 38:69-76.

Chapin, F.S., M.S. Bret-Harte, M. Syndonia, S.E. Hobbie, and H. Zhong. (1996) Plant
functional types as predictors of transient responses of artic vegetation to global
change. Journal of Vegetation Science. 7:347-58.

Chapin, F.S., R.L. Jefferies, J.F. Reynolds, G.R. Shaver, and J. Svoboda. (1992) Arctic
Ecosystems in a Changing Climate: an Ecophysiological Perspective. Academic
Press, Inc. San Diego, Ca. 469 pp.

Chapin, F.S., J.P. McFadden, and S.E. Hobbie. (1997) The role of arctic vegetation in
ecosystem and global processes. In Woodin, SJ ., and M. Marquiss (eds.)
Ecology of Arctic Environment. Blackwell Science. Malden, Ma. p. 97-112.

116

Chapin, F.S., and GR. Shaver. (1985a) Arctic. In Chabot, BE, and H.A. Mooney
(eds.) Physiological Ecology of North American Plant Communities. Chapman
and Hall. New York, NY. p. 16-40.

Chapin, F.S., and GR. Shaver. (1985b) Individualistic growth response of tundra plant
species to environmental manipulations in the ﬁeld. Ecology. 66:564-76.

Chapin, F.S., G.R. Shaver, A.E. Giblin, KJ. Nadelhoffer, and J .A. Laundre. (1995)
Response of Arctic Tundra to experimental and observed changes in climate.
Ecology. 76:696-71 1.

Chapman, W.L., and IE. Walsh. (1993) Recent variations of sea ice and air temperatures
in high latitudes. Bulletin of the American Meteorlogical Society. 74:33-47.

Chaves, M.M., and LS. Pereira. (1992) Water stress, C0; and climate change. Journal
of Experimental Botany. 43: 1 131-39.

Chemov, Y.I. (1985) The Living Tundra. Studies in Polar Research. Cambridge
University Press. Cambridge. 213 pp.

Chester, A.L., and GR. Shaver. (1982) Reproductive effort in cotton grass tussock
tundra. Holarctic Ecology. 5:200-206.

Chitty, D. (1996) Do Lemmings Commit Suicide ?: Beautiful Hypotheses and Ugly Facts.
Oxford University Press. New York, NY. 268 pp.

Christophersen, H., and H. Larcher. (1973) Temperature and Life. Springer-Verlag.
Berlin. 779 pp.

Clements, RE. (1916) Plant Succession: an Analysis of the Development of Vegetation.
Carnegie Institution of Washington Pub]. No. 242. Carnegie Institution of
Washington. Washington, DC. 512 pp.

Culotta, E. (1995) Will plants proﬁt from high C02? Science 268:654-656.
Dahlman, RC. (1993) C02 and plants: revisited. Vegetatio. 104/105:339-55.

Davis, MB. (1989) Lags in vegetation response to greenhouse warming. Climatic
Change. 15:75-82.

Delcourt, P.A., and Delcourt HR. (1981) Vegetation maps for eastern North America:
40,000 BP to the present. In Romans, R.C. (ed.) Geobotany II. Plenum Press.
New York, NY. p. 23-44.

Dennis, JG. (1968) Growth of Tundra Vegetation in Relation to Arctic
Microenvironments at Barrow, Alaska. PhD. Dissertation. Duke University. 288

PP-

117

Dennis, J .G., L.L. Tieszen, and M.A. Vetter. (1978) Seasonal dynamics of above- and
belowground production of vascular plants at Barrow, Alaska. In Tieszen, L.L.
(ed.) Vegetation and Production Ecology of an Alaskan Arctic Tundra. Springer-
Verlag. New York, NY. p. 113-138.

Diaz, 8., and M. Cabido. (1997) Plant functional types and ecosystem function in
relation to global change. Journal of Vegetation Science. 8:463-74.

Dickinson, R.E. (1986) How will climate change? In Bolin, B., B.R. Doos, J. Jafer, and
R.A. Warrick. (eds.) The Greenhouse Effect, Climatic Change, Ecosystems. John
Wittley and Sons. p. 207-70.

Dickinson, R.E., and RJ. Cicerone. (1986) Future global changes from atmospheric
trace gases. Nature. 319:109-15.

Dunbar, MJ. (1968) Ecological Development in Polar Regions: a Study in Evolution.
Prentice-Hall. Englewood Cliffs, NJ. 119 pp.

Dyurgerov, MB, and M.F. Meier. (1997) Year-to-year ﬂuctuations of global mass
balances of small glaciers and their contribution to sea-level changes. Arctic and
Alpine Research. 29:392-402.

Eamus, D., and PG. Jarvis. (1989) The direct effects of increase in the global
atmospheric C02 concentration on natural and commercial temperate trees and
forests. Advanced Ecological Research. 19:1-57.

Emanuel, W.R., H.H. Shugart, and M. Stevenson. (1985) Climatic change and the broad-
scale distribution of terrestrial ecosystem complexes. Climatic Change. 7:29-43.

Gleason, H.A. (1926) The Individualistic Concept of the Plant Association. Bulletin
Torrey Botanical Club. 53:7-26.

Goodwin, C.W., J. Brown, and 8.1. Outcalt. (1984) Potential responses of permafrost to
climatic warming In McBeath, J.H., G.P. Juday, G. Weller, and M. Murray.
(eds.) The Potential Eﬁ’ects of Carbon Dioxide-Induced Climatic Changes in
Alaska. University of Alaska. Fairbanks, Ak. p. 92-105.

Graves, J., and D. Reavey. (1996) Global Environmental Change: Plants, Animals, and
Communities. Longman Group Limited. Harlow, England. 226 pp.

Grime, JP. (1990) Ecological effects of climate change on plant populations and
vegetation. In Jackson, M.T.; G.V. Ford-Lloyd and ML. Parry (eds.), Climate
Change and Plant Genetic Resources. Belhaven Press. London. 60 pp.

Gurevitch, J ., and L.V. Hedges. (1993) Meta-analysis: combining the results of

independent experiments. In S. Scheiner, J .G. (ed.) Design and Analysis of
Ecological Experiments. Chapman and Hall. London. p. 378-98.

118

Hansen, J ., M. Sato, J. Glascoe and R. Ruedy. (1998) A common-sense climate index: is
climate changing noticeably? Proceedings of the National Academy of Science.
95:41 13-4120.

Hansen, K. (1973) Microthermic observations in arctic vegetation. Meddelelser 0m
Gronland. 194:1-32.

Harte, J ., and R. Shaw. (1995) Shifting dominance within a montane vegetation
community: results of a climate-warming experiment. Science. 267:876-79.

Heide, OM. (1983) Physiological aspects of climatic adaptation in plants with special
reference to high-latitude environments. In Kaurin, A., O. J unttia, and J. Nilsen
(eds.) Plant Production in the North: Proceeding of a Workshop Plant
Adaptations. Oslo, Norway. p. 1-22.

Henderson-Sellers, A. (1990) Modeling and Monitoring Greenhouse’ Warming. Trends
in Ecology and Evolution. 5:270-273.

Henderson-Sellers, A., and K. McGufﬁe. (1995) Global models and ’dynamic’ vegetation
changes. Global Change Biology. 1:63-75.

Henry, G.H.R., B. Freedman, and J. Svoboda. (1986) Effects of fertilization on three
tundra plant communities of a polar desert oasis. Canadian Jouranl of Botany.
64:2502-7.

Henry, G.H.R., and U. Molau. (1997) Tundra plants and climate change: the
International Tundra Experiment (ITEX). Global Change Biology. 3: 1-9

Hinkel, K.M., RE. Nelson, Y. Shut, J. Brown, and KR. Everett. (1996) Temporal
changes in moisture content of the active layer and near-surface permafrost at
Barrow, Alaska, U.S.A.: 1962-1994. Arctic and Alpine Research. 28:300-310.

Hinkel, K.M., S.I. Outcalt, and A.E. Taylor. (1997) Seasonal patterns of coupled ﬂow in
the active layer at three sites in northwest North America. Canadian Journal of
Earth Science. 34:664-78.

Hobbie, S.E. (1996) Temperature and plant species control over litter decomposition in
Alaskan tundra. Ecological Monographs. 66:503-22.

Hoffman, A.A., and RA. Parsons. (1997) Extreme Environmental Change and Evolution.
Cambridge University Press. Cambridge. 259 pp.

Hofmann, D.J., J .T. Peterson, and R.M. Rosson. (1996) Climate Monitoring and
Diagnostics Laboratory No. 23 Summary Report [994-95. National Oceanic and
Atmospheric Administration. Environmental Research Laboratories. Boulder, Co.
161 pp.

119

Holdridge, LR. (1947) Determination of world formations from simple climatic data.
Science. 105:367-68.

Holt, RD. (1990) The microevolutionary consequences of climate change. Trends in
Ecology and Evolution. 5:311-15.

Houghton, J .T., G.J. Jenkins, and J .J . Ephraums. (1990) Climate Change. The IPCC
Scientiﬁc Assessment. Cambridge University Press. Cambridge. p. 364 pp.

Houghton, J .T., L.G. Meira Filho, J. Bruce, H. Lee, B.A. Callander, E. Haites, N. Harris,
and K. Maskell. (1996) Climate Change 1995: The Science of Climate Change.
Cambridge University Press. Cambridge. 572 pp.

Houghton, J .T. (1997) Global Warming: The Complete Brieﬁng. Cambridge University
Press. Cambridge. 251 pp.

Huntley, B. (1991) How plants respond to climate change: migration rates, individualism
and the consequences for plant communities. Annals of Botany. 67:15-22.

Huntley, B., and W. Cramer. (1997) Arctic ecosystems and environmental change. In
Crawford, R.M.M. (ed.) Disturbance and Recovery in Arctic Lands: An
Ecological Perspective. Kluwer Academic Publishers. Boston, Ma. p. 1-24.

Jefferies, R.L., J. Svoboda, G. Henry, M. Raillard, and R. Reuss. (1992) Tundra grazing
systems and climatic change. In Chapin, F.S., R.L. Jefferies, J .F. Reynolds, G.R.
Shaver, and J. Svoboda (eds.) Arctic Ecosystems in a Changing Climate.
Academic Press, Inc. San Diego, Ca p. 391-412.

Johnson, P.L. (1969) Arctic plants, ecosystems and strategies. Arctic 22:341-55.

J onasson, S. (1992) Plant responses to fertilization and species removal in tundra related
to community structure and clonality. 0ikos. 63:420-429.

J ones, M.H., C. Bay, and U. Nordenhall. (1997) Effects of experimental warming on
arctic willows (Salix spp.): a comparision of responses from the Canadian High
arctic, Alaskan Arctic, and Swedish Subarctic. Global Change Biology. p. 55-60.

J onsdottir, LS. (1995) "Importance of sexual reproduction in arctic clonal plants and
their evolutionary potential." Global change and Arctic terrestrial ecosystems.
Springer-Verlag. New York, NY. p. 82-88.

Kane, D.L., L.D. Hinzman, and J .P. Zarling. (1991) Thermal response of the active layer
to climatic warming in a permafrost environment. Cold Regions Science and
Technology. 19:1 1 1-22.

120

Kane, D.L., L.D. Hinzman, M. Woo, and KR. Everett. (1992) Arctic hydrology and
climate change. In Chapin, F.S., R.L. Jefferies, J .F. Reynolds, G.R. Shaver, and
J. Svoboda (eds.) Arctic Ecosystems in a Changing Climate: An Ecophysiological
Perspective. Academic Press, Inc. San Diego, Ca.. p. 35-57.

Karl, T.R., R.W. Knight, D.R. Easterling, and G.G. Quayle. (1995) Trends in the U.S.
climate during the twentieth century. Consequences. 1:2-12.

Kattenberg, A., F. Giorgi, H. Grass], G.A. Meehl, J .F.B. Mitchell, R.J. Stouffer, T.
Tokioka, AJ. Weaver, and T.M.L. Wigley. (1996) Climate models - projections
of future climate. In Houghton, J .T., L.G. Meira Filho, B.A. Callander, N.
Harris, A. Kattenberg, and K. Maskell (eds.) Climate Change I 995: The Science
of Climate Change. Cambridge University Press. Cambridge. p. 285-357.

Kennedy, AD. (1995) Simulated climate change: are passive greenhouses a valid
microcosm for testing the biological effects of environmental perturbations?
Global Change Biology. 1:29-42.

Koch, G.W., and H.A. Mooney. (1996) Response of terrestrial ecosystems to elevated
C02. A synthesis and summary. In Koch, G.W. and H.A. Mooney. (eds.)
Carbon Dioxide and Terrestrial Ecosysterrrs. Academic Press. New York. NY.
443 pp.

Khmer, C. (1994) Impact of atmospheric changes on high mountain vegetation. In
Beniston, M. (ed.) Mountain Environments in Changing Climates. Routledge.
New York, NY. p. 155-66.

Khmer, Ch., and W. Larcher. (1988) Plant life in cold climates. In Long, S.P., and El.
Woodward (ed.) Plants and temperature. Symposium of the Society for
Experimental Biology. Company of Biologists. Cambridge. p. 25-57.

Lachenbruch, A.H., and B.V. Marshall. (1986) Changing climate: geothermal evidence
from permafrost in the Alaskan arctic. Science. 234:689-96.

Laird, N .M., and J .H. Ware. (1982) Random-effects models for longitudinal data.
Biometrics. 38:963-74.

Larcher, W. (1975) Physiological Plant Ecology. Springer-Verlag. Berlin. 252 pp.

Leith, H. (1974) Phenology and Seasonality Modeling. Springer-Verlag. Berlin. 444 pp.

Long, S.P. (1991) Modiﬁcation of the response of photosynthetic productivity to rising
temperature by atmospheric C02 concentrations: has its importance been
underestimated? Plant, Cell, and Environment. 14:729-39.

Love, A., and D. Love. (1974) Origin and evolution of the arctic and alpine ﬂoras. In
Ives, J ., and R.G. Barry (eds.) Arctic and Alpine Environments. Methuen.
London. p. 571-603.

121

MacCracken, M.C., M.I. Budyko, A.D. Hecht, and Y.A. Izrael. (1990) Prospects for
Future Climate: A Special US/USSR Report on Climate and Climate Change.
Lewis Publishers, Inc. Chelsea, Mi. 235 pp.

Manabe, S. and RJ. Stouffer. (1979) A climate sensitivity study with a mathematical
model of the global climate. Nature. 282:491-493.

Marion, G.M., G.H.R. Henry, D.W. Freckman, J. Johnstone, G. Jones, M.H. Jones, E.
Levesque, U. Molau, P. Molgaard, A.N. Parsons, J. Svoboda, and R.A. Virginia.
(1997) Open-top designs for manipulating ﬁeld temperature in high-latitude
ecosystems. Global Change Biology. 3:20-32.

Marion, G.M., G.H.R. Henry, P. Molgaard, W.C. Oechel, M.H. Jones, and G. Vourlitis.
(1993) Open-top devices for manipulating ﬁeld temperatures in tundra
ecosystems. In Lunardini, V.J. and S.L. Bowen (eds.) Fourth International
Symposium on Thermal Engineering and Science for Cold Regions. Hanover,
NH. p. 205-210.

Maxwell, B. (1992) Arctic climate: potential for change under global warming. In
Chapin, F.S., R.L. Jefferies, J .F. Reynolds, G.R. Shaver, and J. Svoboda. Arctic
Ecosystems in a Changing Climate. Academic Press. San Diego, Ca. p. 11-34.

McGraw, J .B., and N. Fetcher. (1992) Response of Tundra plant populations to climatic
change. In Chapin, F.S., R.L. Jefferies, J .F. Reynolds, G.R. Shaver, and J.
Svoboda (eds.) Arctic Ecosystems in a Changing Climate: An Ecophysiological
Perspective. Academic Press, Inc. San Diego, Ca. p. 359-76.

Melillo, J .M., A.D. McGuire, D.W. Kicklighter, B.I. Moore, CJ. Vorosmarty, and AL.
Schloss. (1993) Global climate change and terrestrial net primary production.
Nature. 363:234-40.

Microsoft Corporation. (1997) Microsoft Excel 97 Worksheet Function Reference.
Microsoft Corporation. Boston, Ma. p. 370.

Miller, P.C., PJ. Webber, W.C. Oechel, and LL. Tieszen. (1980) Biophysical processes
and primary production. In Brown, J ., P.C. Miller, L.L. Tieszen, and F.L.
Bunnell (eds.) An Arctic Ecosystem: The Coastal Tundra at Barrow, Alaska.
Dowden, Hutchinson & Ross, Inc. Stroudsburg, Pa. 12:66-101.

Molau, U. (1993a) Relationships between ﬂowering phenology and life history strategies
in tundra plants. Arctic and Alpine Research. 25:391-402.

Molau, U. (1993b) IT EX Manual. Danish Polar Center. Copenhagen. p. 29+xv.

Molau, U. (1997) Response to natural climatic variation and experimental warming in
two tundra plant species with contrasting life forms: Cassiope tetragona and
Ranunculus nivalis. Global Change Biology. p. 97-107.

122

Molau, U., and P. Molgaard. (1996) ITEX Manual. Danish Polar Center. Copenhagen.
53 pp.

Molgaard, P. (1982) Temperature observations in high arctic plants in relation to
microclimate in the vegetation of Peary Land, North Greenland. Arctic and
Alpine Research. 14:105-15.

M¢1gaard. P., and K. Christensen. (1997) Response to experimental warming in a
population of Papaver Radicatum in Greenland. Global Change Biology. 3:116-
124.

Monserud, R.A., NM. Tchebakova, and R. Leemans. (1993) Global vegetation change
predicted by the modiﬁed Budyko model. Climatic Change. 25:59-83.

Mueller, J .G. (1996) A Multi-scale 615 Analysis of Active Layer Thiclaress and
Vegetation Type on Alaska 's North Slope. Master’s Thesis. State University of
New York, NY. 132 pp.

Myers, J .P., and EA. Pitelka. (1979) Variations in summer temperature patterns Barrow,
Alaska; analysis and ecological interpretation. Arctic and Alpine Research
11:131-44.

Nadelhoffer, K.J., A.E. Giblin, G.R. Shaver, and J.A. Laundre. (1991) Effects of
temperature and substrate quality on element mineralization in six Arctic soils.
Ecology. 72:242-53.

Nelson, F.E., S.I. Outcalt, G.W. Goodwin, and KM. I-Iinkel. (1985) Diurnal Thermal
Regime in a Peat-Covered Palso, Toolik lake, Alaska. Arctic. 38:310-315.

Oberbauer, S.F., N. Sionit, S.J. Hastings, and W.C. Oechel. (1986) Effects of CO2
enrichment and nutrition on growth, photosynthesis, and nutrient concentration of
Alaskan tundra plant species. Canadian Journal of Botany. 64:2993-98.

Oechel, W.C., T. Callaghan, T. Gilmanov, J.I. Holten, B. Maxwell, U. Molau, and B.
Sveinbjomsson (eds.) (1997) Global change and arctic terrestrial ecosystems.
Ecological Studies, 124. Springer. New York, NY. 493 pp.

Oechel, W.C., A.C. Cook, S.J. Hastings, and G.L. Vorlitis. (1997) Effects of C02 and
climate change on arctic ecosystems. In Woodin, S.J., and M. Marquiss. (eds.)
Ecology of Arctic Environments. Blackwell Science. Malden, Ma. p. 255 - 273.

Oerlemans, J. (1994) Quantifying global warming from the retreat of glaciers. Science.
264:243-45.

Oosting, HJ. (1956) The study of plant communities. W.H. Freeman and Co. San
Francisco, Ca. 440 pp.

123

Ozenda, P. and J .L. Borel. (1989) The possible responses of vegetation to a global
climatic change. In Boer, M.M., and RS. DeGroot. (eds.) Landscape -
Ecological Impact of Climatic Change. IOS Press. Amsterdam. p. 221-49.

Parsons, A.N., J .M. Welker, P.A. Wookey, M.C. Press, T.V. Callaghan, and J .A. Lee.
(1994) Growth responses of four sub-arctic dwarf shrubs to simulated
environmental change. Journal of Ecology. 82:307-18.

Pearcy, R.W., and O. Bjorkrnan. (1983) Physiological effects. In Lemon, E.R. (ed.)
C02 and Plants: The Response of Plants to Rising Levels of Atmospheric Carbon
Dioxide. Westview. Boulder, Co. p. 65-106.

Peters, R.L., and TE. Lovejoy. (1992) Global warming and biodiversity. Yale
University Press. New Haven. 386 pp.

Prentice, C., W. Cramer, S.P. Harrison, R. Leemans, R.A. Monserud, and A.M. Solomon.
(1992) A global biome model based on plant physiology and dominance, soil
properties and climate. Journal of Biogeography. 19:117-34.

Prentice, LC, and MT Sykes. (1995) Vegetation geography and global carbon storage
changes. In Woodwell, G.M. and RT. MacKenzie. (eds.) Biotic Feedbacks in
the Global Climatic System. Oxford University Press. New York, NY. p. 302-12.

Rastetter, ED. (1996) Validating models of ecosystem response to global change.
BioScience. 46:190-198.

Reed, J.C., and AG. Ronhovde. (1971) Arctic Laboratory - a history (1947 - 1966) of
the naval arctic research laboratory at Point Barrow, Alaska. Arctic Institute of
North America. Washington, DC. 748 pp.

Rizzo, B., and E. Wiken. (1992) Assessing the sensitivity of Canada’s ecosystems to
climatic change. Climatic Change. 21:37-55.

Robinson, C. H., and P. A. Wookey. (1997) Microbial ecology, decomposition and
nutrient cycling. In Woodin, S.J., and M. Marquiss. Ecology of Arctic
Environments. Blackwell Science. London. p. 41-68.

SAS Institute. (1990) SAS/STAT User's Guide. SAS Institute. Cary, NC. 890 pp.

SAS Institute. (1996) SAS/ST AT Software: Changes and Enhancement for Release 6. 12.
SAS Institute. Cary, NC. 158 pp.

Savile, D.B.O. (1972) Arctic Adaptations in Plants. Canada Department of Agriculture.
Ottawa, Canada. 81 pp.

124

Schimel, D., LG. Enting, M. Heimann, T.M.L. Wigley. D. Raynaud, D. Alves, and U.
Siegenthaler. (1995) C02 and the carbon cycle. In Houghton, J .T., LG. Meira
Filho, J. Bruce, Hoesung Lee, B.A. Callander, E. Haites, N. Harris, and K.Maskell
(eds.) Climate Change I 994: Radiative Forcing of Climate Change and An .
Evaluation of the IPCC IS92 Emission Scenarios. Intergovernmental Panel on
Climate Change. Cambridge University Press. Cambridge. p. 35-71.

Shaver, GR. (1996) Integrated ecosystems research in northern Alaska 1947-1994. In
Reynolds, J .F., and J .D. Tenhunen (eds.) landscape Function and Disturbance in
Arctic Tundra. Springer-Vedag. New York, NY. p. 19-33.

Shaver, G.R., and F.S. Chapin. (1986) Effect of fertilizer on production and biomass of
tussock tundra, Alaska, U.S.A. Arctic and Alpine Research. 18:261-68.

Shaver, G.R., N. Fetcher, and F.S. Chapin. (1986) Growth and ﬂowering in Eriophorum
vaginatum: annual and latitudinal variation. Ecology. 67 :1524-35.

Shaver, G.R., and J. Kummerow. (1992) Phenology, resource allocation, and growth of
Arctic vascular plants. In Chapin, F.S., R.L. Jefferies, J .F. Reynolds, G.R.
Shaver, and J. Svoboda (eds.) Arctic Ecosystems in a Changing Climate.
Academic Press. San Diego, Ca. p. 193-211.

Shiklomanov, N. (1997) An Analytical Model for Estimating Active-Layer Thickness in
the Kuparuk River Basin, Alaska. Master’s Thesis. University at Albany, State
University of New York. New York, NY. 103 pp.

Shugart, H.H., and TM. Smith. (1996) A review of forest patch models and their
application to global change research. Climatic Change. 34: 131-53.

Smith, M.W. (1975) Microclimatic inﬂuences on ground temperatures and permafrost
distribution, Mackenzei Delta, Northwest Territories. Canadian Journal of Earth
Sciences. 12:421-38.

Solomon, A.M., and A.P. Kirilenko. (1997) Climate change and terrestrial biomass: what
if trees do not migrate? Global Ecology and Biogeography Letters. 6: 139-48.

Solomon, A.M., and H.H. Shugart. (1993) Vegetation dynamics and global change.
Chapman and Hall. New York, NY. 338 pp.

Sorensen, T. (1941) Temperature relations phenology of the northeast Greenland
ﬂowering plants. Meddelelser 0m Gronland. 125:1-305.

Sparks, T.H., and PD. Carey. (1995) The response of species to climate over two
centuries: an analysis of the Marsham phenological record. Journal of Ecology.
83:321-29.

125

 

 

 

Stone, R., T. Mefford, E. Dutton, D. Longenecker, B. Halter, and D. Endres. (1996)
Barrow, Alaska, Surface Radiation and Meteorological Measurements: January
1992 to December 1994. National Oceanic and Atmospheric Administration
Environmental Research Labratories. Boulder, Co. 79 pp.

Stonehouse, B. (1989) Polar Ecology. Tertiary Level Biology. Chapman and Hall. New
York, NY. 222 pp.

Swift, MJ., O.W. Heal, and J .M. Anderson. (eds.) (1979) Decomposition in Terrestrial
Ecosystems: Studies in Ecology Volume 5. University of California Press.
Berkeley, Ca. 372 pp.

Tilman, D. (1987) Ecological Experimentation: Strengths and Conceptual Problems.
Springer-Verlag. New York, NY. 157 pp.

Tissue, D.T., and W.C. Oechel. (1987) Response of Eriophorum vaginatum to elevated
C02, and temperature in the Alaskan Tussock tundra. Ecology. 68:401-10.

Tyrtikov, A.P. (1959) Perennially frozen ground and vegetation. In National Research
Council of Canada (ed.) Principles of Geocriology. National Research Council of
Canada. Ottawa, Canada. p. 399-421.

Waelbroeck, C., P. Monfray, W.C. Oechel, S. Hastings, and G. Vourlitis. (1997) The
impact of permafrost thawing on the carbon dynamics of tundra. Geophysical
Research Letters. 24:229-32.

Walker, DA. (1986) Height and growth rings of Salix lanata ssp. richardsonii along the
coastal temperature gradient of northern Alaska Canadian Journal of Botany.
65:988-93.

Walker, D.A., and PJ. Webber. (1979) Relationships of soil acidity and air temperature
to the wind and vegetation at Prudhoe Bay, Ak. Arctic. 32:224-36.

Walker, L] . (1997) Plant Responses to Emerimental Warming of a Dry Heath Tundra
at Barrow, Alaska. Master’s Thesis. Michigan State University. East Lansing,
Mi. 102 pp.

Walker, MD. (1995) Patterns and causes of Arctic plant community diversity. In
Chapin, F.S. and C. K6mer. (eds.) Arctic and Alpine Biodiversity. Springer-
Verlag. Berlin, Germany. p. 3-20.

Walker, MD. (1996a) A circumpolar comparison of tundra response to temperature
manipulation: a synthesis of international tundra experiment data (absn'act). Arctic
System Science Land-AtmosphereJce Interactions 1996 Science Workshop
Battelle Seattle Conference Center. February 22-24 p. 83-88.

Walker, MD. (1996) Community baseline measurements for ITEX studies. In Molau,
U. (ed.) ITEX Manual. Danish Polar Center. Copenhagen. p. 34 — 36.

126

 

 

 

Walker, M.D., R.C. Ingersoll, and PJ. Webber. (1995) Effects of interannual climate
variation on phenology and growth of two alpine forbs. Ecology. 76: 1067-83.

Warren Wilson, J. (1957) Arctic plant growth. Advancement of Science. 13:383-88.

Watson, R.T., M.C. Zinyowera, and RH. Moss. (1996) Impacts, Adaptations, and
Migration of Climate Change: Scientiﬁc-Technical Analyses. Climate Change.
1995. Cambridge University Press. Cambridge. 878 pp.

Webber, PJ. (1971) Gradient Analysis of the Vegetation Around the Lewis Valley North-
Central Baﬂin Island, Northwest Teritories, Canada. Ph.D. Thesis. Queen’s
University at Kingston. Canada. 366 pp.

Webber, PJ. (1978) Spatial and temporal variation of the vegetation and its production,
Barrow, Alaska. In Tieszen, L.L. (ed.) Vegetation and Production Ecology of
an Alaskan Arctic Tundra. Springer-Vedag. New York, NY. 29:37-112.

Webber, P.J . (1990) Impacts on the terrestrial biological environment: state of
knowledge, speculations and a research agenda. In. McCullock J .A.W. (ed.) The
Symposium on the Arctic and Global Change. The Climate Institute. Washington,
DC. p. 72-80.

Webber, P.J., J .C. Emerick, D.C. Ebert May, and V. Komarkova. (1976) The impact of
increased snowfall on alpine vegetation. In H.W. Steinhoff, and J .D. Ives. (eds.)
Ecological impacts of snowpack augmentation in the San Juan Mountains,
Colorado. CSU-FNR-7052-1. Colorado State University. Fort Collins, Co. p. 201-
264.

Webber, PJ., and MD. Walker. (1991) International Tundra Experiment (ITEX):
Resolution. Arctic and Alpine Research. 23:124.

Weller, G. (1984) A monitoring strategy to detect carbon dioxide-induced climatic
changes in the polar regions. In McBeath, J .H., G.P. Juday, G. Weller, and M.
Murray (eds.) The Potential Eﬁ’ects of Carbon Dioxide-Induced Climatic Changes
in Alaska. University of Alaska. Fairbanks, Ak. p. 23-30.

Wielgolaski, F.E., and L. Karenlampi. (1975) Plant phenology of fennoscandian tundra
areas. In Wielgolaski, F.E. (ed.) F ennoscandian Tundra Ecosystems: Plants and
Microorganisms. Springer-Verlag. New York, NY. p. 94-102.

Woodward, F.I. (1993) The lowland-to-upland transition - modeling plant responses to
environmental change. Ecological Applications. 3:404-8.

Woodward, F.I., and W. Cramer. (1996) Plant functional types and climatic changes.
Journal of Vegetation Science. 7:309-20.

127

 

 

MICHIGAN smrE UNIV. LIBRARIES
m1“WWIWWINW"HMWNI‘IUW
31293016839015