Onl_Er_nph_12647 1..10


Drivers of leaf-out phenology and their implications for species
invasions: insights from Thoreau’s Concord

Caroline Polgar, Amanda Gallinat and Richard B. Primack

Department of Biology, Boston University, Boston, MA 02215, USA

Author for correspondence:
Richard B. Primack
Tel: +1 617 353 2454

Email: primack@bu.edu

Received: 9 September 2013
Accepted: 7 November 2013

New Phytologist (2013)
doi: 10.1111/nph.12647

Key words: chilling requirements, Concord,
Henry David Thoreau, invasive species, leaf
out, Massachusetts, native species,
phenology.

Summary

� To elucidate climate-driven changes in leaf-out phenology and their implications for species
invasions, we observed and experimentally manipulated leaf out of invasive and native woody

plants in Concord, MA, USA.
� Using observations collectedby HenryDavidThoreau (1852–1860) andour own observations
(2009–2013), we analyzed changes in leaf-out timing and sensitivity to temperature for 43

woody plant species. We experimentally tested winter chilling requirements of 50 species by

exposing cut branches to warm indoor temperatures (22°C) during the winter and spring of
2013.
� Woody species are now leafing out an average of 18 d earlier than they did in the 1850s, and
are advancing at a rate of 5 � 1 d °C�1. Functional groups differ significantly in the duration of
chilling they require to leaf out: invasive shrubs generally have weaker chilling requirements than

native shrubs and leaf out faster in the laboratory and earlier in the field; native trees have the

strongest chilling requirements.
� Our results suggest that invasive shrub species will continue to have a competitive advantage
as the climate warms, because native plants are slower to respond to warming spring

temperatures and, in the future, may not meet their chilling requirements.

Introduction

The connection between plant phenology and temperature is well
established for temperate ecosystems; spring plant phenophases,
such as leaf out and flowering, generally occur earlier in warmer
years (Cleland et al., 2007; Polgar & Primack, 2011). Across
temperate regions of the world, woody species are advancing their
timing of leaf out at an average rate of 4.6 d °C�1 in response to
warming annual temperatures (Wolkovich et al., 2012). However,
the response of leafing out to warming is species specific
(Lechowicz, 1984; Cleland et al., 2007), and the magnitude of
the response of individual species to temperature can affect species
composition at the community level. Weak responses to temper-
ature in flowering are linked to native species declines (Willis et al.,
2008) and strong responses are linked to increases in abundance of
non-native species (Hulme, 2011). Strong responses of leafing out
to increased temperature may increase the growing season length of
a species and its ability to compete for local resources in spring
(Fridley, 2012).

As spring temperatures have reached record highs, flowering
and leaf-out dates continue to advance in many temperate areas
(Wolkovich et al., 2012; Ault et al., 2013; Ellwood et al., 2013).
Leaf-out models that consider only heat sums, the accumulation
of warm temperatures (or thermal units) that force bud burst,
predict continued advances in leaf-out dates as winter and spring
temperatures warm (Vitasse et al., 2009). However, before the

accumulation of heat sums, temperate plants must accumulate a
specific number of chilling units for the release from dormancy,
referred to as the chilling requirement (Falusi & Calamassi,
1990; Faust et al., 1997). As winter and spring temperatures
continue to warm, the fulfillment of chilling requirements may
be delayed or may go unmet for many species, possibly delaying
spring onset or causing abnormal leaf-out events, such as
extremely late bud burst or incomplete leaf development
(McCreary et al., 1990; Morin et al., 2009; Fu et al., 2012). In
these situations, plants may need additional warming to
overcome the unmet winter chilling requirement. In fact, delays
in spring leafing out attributed to incomplete winter chilling have
already been reported in some regions, such as the southeastern
USA and the Tibetan Plateau (Zhang et al., 2007; Yu et al., 2010;
Cook et al., 2012).

Chilling requirements are believed to serve as a way to protect
plants from leafing out too early in the spring in response to brief
warm spells, when they would still be vulnerable to freezing events
after the warm spell had passed (Gu et al., 2008; Korner & Basler,
2010; Basler & Korner, 2012). Winter chilling, spring warming
and photoperiod have all been identified as requirements necessary
forspring leafout.Althoughseveralstudiessuggest thatchillingand
warming have a stronger effect than photoperiod on leaf-out
phenology, their relative importance and differences among species
remain largely unknown (Vitasse et al., 2009; Polgar & Primack,
2011; Laube et al., 2013; Vitasse & Basler, 2013).

� 2013 The Authors
New Phytologist � 2013 New Phytologist Trust

New Phytologist (2013) 1
www.newphytologist.com

Research



Invasive species often leaf out earlier than native species (Xu
et al., 2007), a difference that suggests that invasive species may
have weaker winter chilling requirements or stronger responses to
spring warming. If, in general, invasive species have significantly
weaker winter chilling requirements or stronger responses to spring
warming than native species, there is a risk that invasive species will
gain an additional competitive advantage as the climate continues
to warm. Differences in response rates of invasive vs native woody
plants may have important implications for the future of temperate
forests, altering species composition, ecosystem functions and
feedbacks to the climate system.

In this study, we investigated the sensitivity of spring leaf-out
phenology to temperature and the relative chilling requirements for
a variety of woody species in Concord and nearby areas of
Massachusetts, USA. Species included native trees and shrubs, non-
native invasive shrubs and vines. We experimentally determined
the duration of chilling required by each species before leaf out and
explored differences between native and invasive species. To study
sensitivity to spring and winter temperatures in the field, we used
observational leaf-out data from Concord from two time periods,
1853–1860 and 2009–2013. We investigated the following
questions. How have first leaf-out dates changed from the 1850s
to the present? Do native trees, native shrubs and non-native
invasive shrubs differ in the duration of chilling they require before
leafing out? How do winter chilling requirements affect the timing
of leaf out in the field and the phenological response of plants to
warmer spring temperatures? The answers to these questions will
ultimately help ecologists predict which woody species will benefit
most and how the temperate forest community will change as
temperatures continue to warm as a result of climate change.

Materials and Methods

Observational field study

ThefieldstudywasconductedinConcord,MA,northeasternUSA,
for three reasons. First, this temperate forest ecosystem has four
distinct seasons, and pronounced winters with multiple days below
freezing and potential freezes into late spring. Second, the weather
in this region of the USA is highly variable both among and within
years. Third, we have extensive prior knowledge of the effects of
climate change on the flowering phenology and abundance of plant
species over the past 150 yr in this location (Miller-Rushing &
Primack, 2008; Willis et al., 2008, 2010; Primack et al., 2009;
Primack & Miller-Rushing, 2012; Ellwood et al., 2013).

We obtained historical records of first leaf dates (FLDs) for
woody plant species in Concord, MA, USA, recorded by the
environmental writer and philosopher Henry David Thoreau from
the Morgan Library in New York City. These observations were
made from 1852 to 1860, with the most complete observations
made in the springs of 1854, 1855 and 1860. From this list, we
selected the 43 deciduous species that were still present in Concord
and for which there were > 2 yr of leaf-out data from the 1800s
(Table 1, Supporting Information Table S2). Thoreau’s observa-
tions of flowering times have already been extensively studied and
compared with modern observations (Primack & Miller-Rushing,

2012; Ellwood et al., 2013), but this is the first scientific use of his
leaf-out observations. Thoreau did not describe his field methods
for recording leaf-out dates. However, Thoreau’s detailed descrip-
tions in his journals of the small size of the young leaves and his
frequent journal entries suggest that his methodology was similar to
ours. We also know that Thoreau was recording the first flowering
date anywhere within Concord, and so it seems likely that he was
recording the first leaf-out date of particular tree species anywhere
in Concord, rather than focusing on specific individuals or plots of
land, again just as we did. In any case, because of the extremely fast
process of leaf flush in Massachusetts, even variations of Thoreau’s
method would almost certainly yield similar, if not identical, values
to our methods.

To obtain the current FLDs of these same 43 species, we visited
Concord twice a week throughout March, April and May of 2009–
2013,andrecorded thefirsttimeeachofthestudyspecieswasinleaf
(Tables 1, S2). We considered a species to have leafed out when at
least three branches on one individual plant anywhere in Concord
had at least one fully unfolded leaf, the definition used by Project
Budburst (www.budburst.org), and similar to that used by the USA
National Phenology Network (www.usanpn.org). We obtained air
temperatures from the Blue Hill Meteorological Observatory in
Milton, MA (data available from the National Climatic Data
Center, www.ncdc.noaa.gov), a site 33 km southeast of Concord,
where temperatures are highly correlated with those in Concord
(Miller-Rushing & Primack, 2008). Of the 43 species, 23 had
complete data for the years 1854, 1855, 1860 and 2009–2013.

In 2013, we recorded the FLDs in the field for an additional 13
species from Concord, and also from Newton (22 km southwest of
Concord) and the Arnold Arboretum of Harvard University
(34 km southwest of Concord; Table 1).

Laboratory chilling study

In 2013, we collected dormant twigs of 50 species of woody plants,
including trees, shrubs and vines. The species collected included 36
ofthe43speciesfrom theobservationalfield studyinConcord,plus
four additional native species and 10 additional invasive species
(Table 1). We collected most twigs from the wild in Newton and
Concord, MA. In some cases, plant species were too rare locally for
us to collect twigs from wild individuals; in those cases, we collected
twigs from plants growing at the Arnold Arboretum (Table 1).

We collected twigs on 8–11 January, 14–20 February, 20–24
March and 18–20 April, 2013. In Table S1, we provide the chilling
days and thermal units that plants had accumulated in the field
before these collection dates. In the first collection period, we
collected 48 species. In February, we did not collect the nine species
that leafed out in fewer than 21 d after being collected in January.
However, in February, we began collecting two additional species,
witch hazel (Hamamelis virginiana) and spicebush (Lindera
benzoin), that we had not collected in January. We collected a
totalof41species inFebruary. InMarch,wecollected all50species.
In April, we collected only those species that took longer than 40 d
to leaf out in the laboratory after the March collection – we wanted
to confirm that they would be able to leaf out under laboratory
conditions when collected as twigs. We collected 20 twigs from

New Phytologist (2013) � 2013 The Authors
New Phytologist � 2013 New Phytologist Trustwww.newphytologist.com

Research

New
Phytologist2



each of the selected species at each collection time. For wild species,
we collected the 20 twigs of each species from at least five
individuals. For species collected at the Arnold Arboretum, we were
only able to collect from one or two individuals per species. In the
analysis described below, we used the twigs to generate a single

leaf-out date for each species in each treatment, and did not treat
them as replicates.

We brought the twigs into a laboratory at Boston University,
where we re-cut the base of each one and placed them into
containers with 500 ml of tap water. The twigs of each species were

Table 1 Species used in the study listed with their functional group, chilling requirement, days to leaf out after twig collection in March 2013, date of leaf out in
the field in 2013 and collection site – Newton (N), Arnold Arboretum (A) and Concord (C)

Common name Scientific name Functional group
Chilling
requirement

Days to leaf
out in March

Date of leaf
out in 2013

Collection
site

Japanese barberry Berberis thunbergii Invasive shrub Minimal 8 9 April N
European barberry1 Berberis vulgaris Invasive shrub Minimal 16 16 April A
Bittersweet Celastrus orbiculatus Invasive vine Minimal 22 2 May N
Silverling Eleagnus umbellata Invasive shrub Minimal 9 14 April A
Compact privet Ligustrum compactum Invasive shrub Minimal 9 9 April A
Ibota privet Ligustrum ibota Invasive shrub Minimal 9 14 April A
Late honeysuckle Lonicera maackii Invasive shrub Minimal 9 9 April A
Subsessilis honeysuckle Lonicera subsessilis Invasive shrub Minimal 9 14 April A
Multiflora rose Rosa multiflora Invasive shrub Minimal 8 9 April N
American elder1 Sambucus canadensis Native shrub Minimal 9 16 April A
Apple1 Malus domestica Non-native tree Minimal 8 9 April N
Winged euonymus Euonymus alatus Invasive shrub Mild 15 24 April A
Black chokeberry1 Aronia arbutifolia Native shrub Mild 9 16 April A
Buttonbush1 Cephalanthus occidentalis Native shrub Mild 16 10 May N
Sweet fern1,2 Comptonia peregrina Native shrub Mild 15 27 April N
Silky dogwood1 Cornus amomum Native shrub Mild 15 24 April N
Hazelnut1 Corylus americana Native shrub Mild 13 24 April C
Sheep’s laurel1,2 Kalmia angustifolia Native shrub Mild 22 27 April N
Bayberry1 Myrica pensylvanica Native shrub Mild 16 10 May A
Meadow sweet1 Spiraea latifolia Native shrub Mild 9 16 April A
Early low-bush blueberry1 Vaccinium angustifolium Native shrub Mild 16 16 April A
Arrowwood1 Viburnum recognitum Native shrub Mild 13 21 April C
Grape1 Vitis aestivalis Native vine Mild 15 2 May N
Paper birch1 Betula papyrifera Native tree Mild 9 24 April A
Gray birch1 Betula populifolia Native tree Mild 15 27 April N
American ash1,2 Fraxinus americana Native tree Mild 33 2 May A
Buckthorn Rhamnus frangula Invasive shrub Moderate 15 27 April N
Wild alder1,2 Alnus serrulata Native shrub Moderate 20 18 April C
Coast pepperbush1,2 Clethra alnifolia Native shrub Moderate 15 27 April N
Huckleberry1 Gaylussacia baccata Native shrub Moderate 15 2 May N
Witch hazel Hamamelis virginiana Native shrub Moderate 15 27 April C
Mountain laurel1,2 Kalmia latifolia Native shrub Moderate 33 7 May A
Staghorn sumac Rhus typhina Native shrub Moderate 15 2 May N
Highbush blueberry1 Vaccinium corymbosum Native shrub Moderate 8 24 April N
Late low-bush blueberry1 Vaccinium pallidum Native shrub Moderate 22 21 April N
Red maple1,2 Acer rubrum Native tree Moderate 15 21 April N
Black birch1 Betula lenta Native tree Moderate 22 27 April N
Pignut hickory1,2 Carya glabra Native tree Moderate 16 2 May A
Bigtooth aspen1,2 Populus grandidentata Native tree Moderate 32 2 May N
Black cherry1,2 Prunus serotina Native tree Moderate 16 16 April A
White oak1,2 Quercus alba Native tree Moderate 22 2 May N
Red oak1,2 Quercus rubra Native tree Moderate 15 27 April N
American elm1,2 Ulmus americana Native tree Moderate 23 24 April A
Spicebush2 Lindera benzoin Native shrub Strong 42 2 May C
Greenbriar1,2 Smilax rotundifolia Native vine Strong 42 10 May N
Silver maple1,2 Acer saccharinum Native tree Strong 43 2 May A
Sugar maple1,2 Acer saccharum Native tree Strong 43 27 April A
American beech1,2 Fagus grandifolia Native tree Strong 42 27 April N
Black gum1,2 Nyssa sylvatica Native tree Strong 42 10 May N
Sassafras2 Sassafras albidum Native tree Strong 40 7 May C

1Species that both we and Thoreau monitored in Concord.
2Species that did not leaf out within 40 d after the January collection.

� 2013 The Authors
New Phytologist � 2013 New Phytologist Trust

New Phytologist (2013)
www.newphytologist.com

New
Phytologist Research 3



evenly split into two groups of 10, and each group was randomly
placed in one of two light banks fitted with plant grow lights
(Sylvania 20 W T12 Florescent Bulb, Gro-Lux Wide Spectrum
Phosphor; Osram Sylvania Ltd, Danvers, MA, USA). Plants in
both light banks were exposed to 14 h of daylight, typical of mid-
April in eastern Massachusetts, to standardize photoperiod
throughout the experiment and to ensure that it was not a factor
in leaf-out timing. The laboratory was maintained at a constant
temperature of 22°C, a temperature selected to ensure rapid
accumulation of thermal units and to be easily repeated by other
researchers. Because the laboratory was well ventilated, the bulbs
did not appreciably affect plant temperature.

We evaluated twigs once each week for leaf out. We considered a
twig to have leafed out when at least one leaf on the stem was fully
unfolded and the top surface of the leaf was visible. We considered a
species to have leafed out when over half of the live twigs in one of
the two light banks had leafed out. In general, species leafed out at
the same time in each of the two light banks and all of the twigs in a
light bank leafed out at about the same time. Occasionally, twigs of
certain species wilted or stopped developing; however, this
occurred rarely and only after they had met our definition of
having leafed out, and so these species were still included in all
analyses.

Each week, we re-cut the base of each stem to expose fresh tissue
and maintain the ability of the twigs to take up water. We changed
the water in each container weekly and washed the twigs to remove
any mold growing on the stem. Weekly twig evaluations and
washing continued for 4–6 wk following each collection date, after
which we discarded the remaining twigs. We calculated the time
between collection and leaf-out dates for each species after each
collection period.

During the winter of 2011–2012, we carried out a pilot study to
examine the feasibility of using cuttings from wild plants to test
chilling requirements, aswellasphotoperiodrequirements.During
the pilot study, we used the same methodology as the 2013 main
chilling study with the following exceptions. The pilot study
included only 17 species, many of which were included in the main
study (Table S2), and we collected twigs in November in addition
to January–April.We exposed 10twigs of each species to one of two
light treatments: ambient day length or extended day length
(14 h : 10 h, day : night).WeidentifiedFLD asthedate onwhich at
least one leaf on one twig was unfolded. American beech (Fagus
grandifolia) was the only species for which photoperiod had an
effect on leaf out. Thus, we did not test photoperiod requirements
in the main study and focused only on chilling requirements.

Analysis

All analyses were performed using JMP Pro 10 statistical software
(SAS Institute Inc., Cary, NC, USA) and an a value cut-off for
significance of 0.05. Our statistical analyses are based primarily on
direct measurements, such as the dates of collection, the number of
days to leaf out and the average spring temperatures, rather than
degree-daymodels,althoughthetwoarehighlycorrelated(Archetti
et al., 2013; Ellwood et al., 2013). We report chilling days and
thermal units of warming in Table S1 for twig collection dates; we

found that thermal units that had accumulated by the date of leaf
out in the field were strongly correlated with dates of leaf out in
2013 (r = 0.97; P < 0.0001).

Field study For the 23 species for which we had complete data for
the years 1854, 1855 and 1860, and 2009–2013, we used a two-
wayANOVA tocomparethedifferences inFLD amongspeciesand
between time periods (Thoreau’s years vs recent years). For each of
the 43 species for which we had multiple years of observations, we
used simple linear regression to test the relationship between FLD
for each year with the mean March and April temperature for that
year. We considered the slope of this relationship to be the
‘sensitivity to temperature’. We then used linear regression to test
the relationship between sensitivity to temperature and the mean
FLD from 2009 to 2013 to determine whether sensitivity to
temperatureisrelatedtotherelative timeduringthespringinwhich
a species typically leafs out (early or late in the season).

Laboratory chilling study We assigned each species to one of four
categories of chilling requirement: minimal, mild, moderate or
strong. These groupings were based on the differences between the
number of days it took for a species to leaf out after being collected
and brought into the laboratory (Table 2). For each collection date,
we calculated the number of chilling days that had occurred before
that date – chilling days were equal to the number of days after 1
November on which the mean temperature was below 5°C
(Murray et al., 1989). In 2013, species in the minimal chilling
requirement category required fewer than 49 winter chilling days
before they were competent to leaf out, species in the mild category
required fewer than 83 winter chilling days, species in the moderate
category required fewer than 114 winter chilling days and species in
the strong category required 114 or more chilling days before they
became competent to leaf out (Table S1). For the pilot study of
2011–2012, winter temperatures were milder, and so species
accumulated fewer chilling days in the field before they were
collected.

We used Spearman rank correlations in analyses for some of our
categoricaldataandfordatathatwerenotnormallydistributed.For
example, we used Spearman rank correlations to determine the
strength of correlation between chilling categories assigned in the
pilot study of 2011–2012 and the main study in 2013 for
overlapping study species.

Connecting field observations and chilling results Based on their
ecological characteristics, we assigned 46 of the 50 species used in
the chilling study to a functional group category: invasive shrub (10
species), native shrub (20 species) or native tree (16 species)
(Table 1). To characterize invasive shrubs, we used the current (as
of August 2013) Invasive Plant Atlas of New England (IPANE) list
(www.eddmaps.org/ipane). Although they are not on the IPANE
invasive plant list, we classified two Ligustrum species (L. compac-
tum and L. ibota) and one Lonicera species (L. subsessilis) as invasive,
because of relatedness and morphological similarity to the many
invasive species in the same genera. Four species did not fit into the
three main functional group categories, and were omitted from the
functional group analyses – two native vines, grape (Vitis aestivalus)

New Phytologist (2013) � 2013 The Authors
New Phytologist � 2013 New Phytologist Trustwww.newphytologist.com

Research

New
Phytologist4



and greenbriar (Smilax rotundifolia), the invasive non-native vine
bittersweet (Celastrus orbiculatus) and the non-native apple tree
(Malus domestica).

We used one-way ANOVA to test for the effects of functional
groups on leaf-out time in March in the laboratory and FLD in the
field. We followed this test with a multiple range test, Tukey’s
honestly significant difference test, to determine which particular
functional groups were different from each other. In order to
separate the effects of functional group and chilling requirement
category on days to leaf out in March and FLD in the field, we used
two-way ANOVA to test for relationships of chilling categories and
functional groups on the variables of days to leaf out in March, field
FLD and sensitivity to temperature.

Results

Field study

Woody plants in Concord are leafing out significantly earlier now
(2009–2013) than they were in Thoreau’s years (1854–1855,

1860). In a two-way ANOVA, there was a significant difference in
FLD among species (P < 0.001), as well as between time periods
(P < 0.001). For the 23 species for which we have complete data for
years 1854, 1855, 1860 and 2009–2013, the mean FLD in
Thoreau’s years was May 8 � 7 d (� SD), whereas the mean leaf
out in recent years was April 20 � 7 d, a difference of 18 d. All
species are leafing out earlier now than in Thoreau’s time
(P < 0.001; Fig. 1). In 2013, there were 31 d between the FLD of
the earliest and latest of the 23 species (Table 1). We found a
significant correlation between order of FLD in the two time
periods, meaning that the species tended to leaf out in the same
sequence in both time periods (r = 0.76; P < 0.001; Fig. 1).

Warmer late-winter and early-spring temperatures in March and
April were associated with earlier FLDs. Winter temperature in
January and February had no apparent effect on FLD, presumably
because there were always sufficient chilling days in Concord by the
time of spring leaf out. The mean March and April air temperature
for Thoreau’s years (1854, 1855, 1860) was 2.8°C; the mean for
these years is representative of the mid-1800s overall, as the mean
March and April temperature for the years 1845–1865 was 2.9°C

Table 2 Definitions of categories of chilling requirements in our pilot experiment and main experiment

Chilling
requirement Pilot experiment Main experiment

Minimal Days to first leaf after November collection ≤ double the days
to first leaf after March collection1

Days to first leaf after January collection within 10 d of days to
first leaf after March collection

Mild Days to first leaf after January collection ≤ double the days
to first leaf after March collection1

Days to first leaf after February collection within 10 d of days to
first leaf after March collection

Moderate Days to first leaf after February collection ≤ double days
to first leaf after March collection

First leaf < 40 d after March collection

Strong Days to first leaf after February collection more than double
days to leaf out after March collection

First leaf > 40 d after March collection

1Also includes species that leafed out within 1 wk of collection in the pilot study for the November and January collections.

Fig. 1 Mean first leaf dates (FLDs) of
individual species for recent years regressed
against the mean FLDs for Thoreau’s years.
The data included in this figure are those for
which we have complete data for the years
1854, 1855, 1860 and 2009–2013. The
relationship between the two variables is
highly significant (slope = 0.69; R2 = 0.62;
P < 0.001). All of the study species are leafing
earlier in our years than they did in Thoreau’s
years. The red line shows the line along which
the points would fall if first leaf dates had not
changed; the black line is the regression line
through the points.

� 2013 The Authors
New Phytologist � 2013 New Phytologist Trust

New Phytologist (2013)
www.newphytologist.com

New
Phytologist Research 5



(Fig. S1). The mean March and April temperature for 2009–2013
was 5.1°C (Fig. S1). There was a significant negative relationship
between mean March and April temperature and FLD for all study
species (Table S3). Among the study species, the sensitivity of FLD
to temperature ranged from �2.8 � 7.3 d °C�1 for silver maple
(Acer saccharinum) to �7.1 � 15.2 d °C�1 for late lowbush
blueberry (Vaccinium angustifolium). The mean response to
temperature was �5.0 � 1.0 d °C�1; that is, for each 1°C warm-
ing, plants leafed out c. 5 d earlier on average. We also found a
significant relationship between sensitivity to temperature and
mean FLD; species that were most sensitive to spring temperatures
tended to leaf out earlier in the spring (P = 0.038).

Laboratory chilling study

Inthepilotstudy,16ofthe17species tookthesame numberofdays
toleafout in each of thetwo day length treatments, indicating alack
of photoperiod response. For the American beech, twigs collected
in November, January and February and placed in the ambient day
length treatment never leafed out in the laboratory, presumably
because this species had strong winter chilling and photoperiod
requirements. However, beech twigs collected at the same time and
kept under extended day conditions did leaf out after more than a
month. For samples collected on 9 March 2012, the last date of
collection in the pilot study, twigs leafed out after 28 d in the
extended day length treatment and after 32 d in the ambient day
length treatment.

There was a significant difference among species and among
collection periods in the number of days it took for leaf out to occur
for both the pilot study in 2011–2012 (Table S2) and the main
study (Table 1, Fig. 2), as determined by ANOVA (P < 0.001 in all
cases). In each case, the number of days required to leaf out declined
with successive collection dates (Fig. 2).

Overall, winter chilling requirement categories in the main study
were minimal for 11 species, mild for 15 species, moderate for 17
species and strong for seven species (Table 1). A Spearman rank
correlation showed a significant correlation between the winter
chilling requirement categories as determined in the pilot study and
the main study (r = 0.75; P = 0.001) despite changes in method-
ology between the 2 yr. For example, in both studies, multiflora
rose (Rosa multiflora) and privet (Ligustrum compactum) were in the
minimal chilling requirement category, whereas American beech
and spicebush were in the strong chilling requirement category.
Although the results of the pilot and main study were correlated,
there was some variation that may be attributed to the change in
definitions of the chilling requirement categories.

Functional groups (e.g. invasive shrubs, native shrubs and native
trees) largely fell into different chilling categories, with invasive
shrubs requiring mostly minimal winter chilling and native trees
having moderate or strong winter chilling requirements (Fig. 3;
Table 1). Native shrubs largely had mild and moderate chilling
requirements. Species with minimal chilling requirements
included eight invasive shrubs, one non-native tree (apple), one
invasive vine (bittersweet) and one native shrub (American elder,
Sambucus canadensis). Species with mild chilling requirements
included primarily native shrubs. Species with moderate chilling

requirements included a mix of trees and native shrubs, although
the group also included one invasive shrub (buckthorn, Rhamnus
frangula). Species with strong chilling requirements were all native
trees, except for one native shrub (spicebush) (Table 1).

Because a majority of species had met their winter chilling
requirements by March, we used the number of days required to leaf
out following March collectionas a metric for species responsiveness
to spring warming as measured in the laboratory. Using one-way
ANOVA, functional groups had significantly different times to leaf
out in March (P < 0.001), with native trees leafing out significantly
later (27 d) than invasive shrubs (11 d) and native shrubs (17 d)
(Table 3; Fig. 2). Two-way ANOVA showed a significantdifference
among chilling requirement categories in the number of days to leaf
out (P < 0.001), but no significant difference among functional
groups (P = 0.34), indicating that invasive species leaf out quickly
largely as a result of their minimal and mild chilling requirements.
Species with minimal chilling requirements were the fastest to leaf
out in March (mean = 11 � 4 d), whereas species with moderate
chilling requirements were the slowest (mean = 19 � 6 d), with
species with mild chilling requirements in between (mean =
15 � 6 d). Species with strong chilling requirements that had not
met their winter chilling requirements by March (as indicated by
requiring > 40 d to leaf out in the laboratory) were excluded from
this analysis; these species included American beech, silver maple,
sugarmaple(Acersaccharum),spicebush,blackgum(Nyssasylvatica),
sassafras (Sassafras alba) and greenbriar (Smilax rotundifolia).

Connecting field observations and chilling results

One-way ANOVA indicated that functional groups differed
significantly (P < 0.001) in their FLDs in the field – invasive
shrubs leafed out significantly earlier (mean = April 15) than native
shrubs (26 April) and native trees (28 April; Table 3). Two-way
ANOVA indicated a significant association between FLD in the
fieldin2013andchillingrequirementcategoryasdeterminedinthe
laboratory (P = 0.005), but not with functional group (P = 0.87).
The species with minimal chilling requirements, as determined in
the laboratory, were the first to leaf out in the field, whereas species
with moderate and strong chilling requirements leafed out later
(Fig. 4). The lack of a significant effect of functional group in the
two-wayANOVAsuggeststhattheassociation offunctionalgroups
with FLDs is through their correlation with chilling categories.

Spearman rank correlation indicated that the order in which the
50 species leafed out in March in the laboratory was strongly related
to the order in which they leafed out in the field in 2013 (r = 0.64;
P < 0.001; Table 1), indicating that clipped twigs serve as a valid
proxy for leaf-out phenology of wild plants. Apple, Japanese
barberry (Berberis thunbergii) and multiflora rose all took 8 d – the
fewest days of any species – to leaf out in the laboratory after the
March collection,and shared theearliest leaf-out date inthefield on
9 April. Sassafras, greenbriar and black gum, however, took at least
40 d to leaf out in the laboratory after the March collection and
leafed out in the field between 6 and 9 May in 2013, among the
latest dates of any species in our study (Table 1).

The sequence of leaf out across species was largely the same in the
field in 2013 as it was in the laboratory after January collection

New Phytologist (2013) � 2013 The Authors
New Phytologist � 2013 New Phytologist Trustwww.newphytologist.com

Research

New
Phytologist6



(Spearman rank correlation r = 0.66; P < 0.001). This suggests that
an extremely warm winter and early spring would not substantially
affect the sequence in which species leaf out. The exceptions were
black cherry (Prunus serotina) and wild alder (Alnus serrulata),
which took > 40 d to leaf out in the laboratory after being collected
in January, but which leafed out relatively early in the field (16 and
18 April, respectively). Bayberry (Myrica pensylvanica) and
buttonbush (Cephalanthus occidentalis) were also exceptions and
leafed out in 30 and 31 d after January collection, but did not leaf
out in the field until relatively late, both on 10 May.

For the subset of 43 species monitored by Henry David Thoreau
and for which we have sensitivity to temperature in the field as
calculated using field observations from the 1850s and the present,
neither chilling requirement category nor functional group
explained the sensitivity to temperature (two-way ANOVA;
P = 0.61; Table S3). However, this dataset excludes all of the
invasive species, with the exception of European barberry (Berberis

vulgaris), because Thoreau did not monitor the leafing out times of
these species.

Discussion

Changes in timing of leaf out

Our study is unusual in exploring links between changing leaf-out
times in the field and winter chilling requirements using laboratory
studies. The most significant results are as follows: all of the species
examined are leafing out earlier now than they did 160 yr ago
because of warmer spring temperatures; non-native invasive species
leaf out earlier in the spring and have weaker winter chilling
requirements than native trees and shrubs; and the leaf-out dates of
invasive shrubs will probably continue to advance faster than those
of native trees and shrubs as winter chilling is reduced and spring
temperatures continue to warm.

B
. t

hu
 

R
. m

ul
 

V
. c

or
 

A
. m

el
 

B
. p

ap
 

E
. u

m
b 

L.
 c

om
 

L.
 ib

o 
L.

 m
aa

 
L.

 s
ub

 
S.

 c
an

 
S.

 la
t 

C
. a

m
e 

V
. r

ec
 

A
. r

ub
 

B
. p

op
 

C
. a

ln
 

C
. p

er
 

C
. a

m
o 

E
. a

la
 

G
. b

ac
 

H
. v

ir
 

Q
. r

ub
 

R
. f

ra
 

R
. t

yp
 

B
. v

ul
 

C
. g

la
 

C
. o

cc
 

M
. p

en
 

P
. s

er
 

V
. a

ng
 

A
. s

er
 

B
. l

en
 

K
. a

ng
 

Q
. a

lb
 

V
. p

al
 

U
. a

m
e 

P
. g

ra
 

F
. a

m
e 

K
. l

at
 

S.
 a

lb
 

F
. g

ra
 

L.
 b

en
 

N
. s

yl
 

A
. s

ac
1 

A
. s

ac
2 

T
he

rm
 u

ni
ts

 to
 M

ar
la

b 
le

af
 o

ut

Order of March lab leaf out

0
100
200
300
400
500
600
700
800

B. thu R.
mul

V. cor A. mel B.
pap

E.
umb

L.
com

L. ibo L.
maa

L. sub S. can S. lat C.
ame

V. rec B.
pop

C.
amo

G.
bac

H. vir R. fra R. typ B. vul C. occ M.
pen

V. ang B. len V. pal

T
he

rm
 u

ni
ts

 to
 J

an
la

b 
le

af
 o

ut

Order of March lab leaf out

0
50

100
150
200
250
300

B
. t

hu
 

R
. m

ul
 

V
. c

or
 

A
. m

el
 

B
. p

ap
 

C
. a

lt 
E

. u
m

b 
L.

 c
om

 
L.

 ib
o 

L.
 m

aa
 

L.
 s

ub
 

S.
 c

an
 

S.
 la

t 
C

. a
m

e 
V

. r
ec

 
A

. r
ub

 
B

. p
op

 
C

. a
ln

 
C

. p
er

 
C

. a
m

o 
G

. b
ac

 
H

. v
ir

 
Q

. r
ub

 
R

. f
ra

 
R

. t
yp

 
B

. v
ul

 
C

. g
la

 
C

. o
cc

 
M

. p
en

 
P

. s
er

 
V

. a
ng

 
A

. s
er

 
B

. l
en

 
K

. a
ng

 
Q

. a
lb

 
V

. p
al

 
P

. t
re

 
R

. c
an

 
U

. a
m

e 
P

. g
ra

 
F

. a
m

e 
K

. l
at

 
S.

 a
lb

 
F

. g
ra

 
L.

 b
en

 
N

. s
yl

 
A

. s
ac

1 
A

. s
ac

2 
M

. g
al

 
R

. v
er

 

T
he

rm
 u

ni
ts

 to
 le

af
ou

t i
n 

fi
el

d 

Order of March lab leaf out

0
100
200
300
400
500
600
700
800

(a)

(b)

(c)

Fig. 2 The number of thermal units of warming to which the twigs of a species had been exposed on the day on which they leafed out in 2013, for (a) the twigs
collectedin Januaryof the laboratoryexperiment, (b) thetwigscollected inMarch of the laboratoryexperiment and (c) the plantsgrowingoutside(see Table 1).
Along the x-axis, species are listed in the order in which the twigs collected in March leafed out in the laboratory for the first time. For species that leafed out for
the first time on the same day, the rankings were assigned alphabetically. The abbreviations stand for the first letter of the genus and first three letters of the
species; A. sac1 is Acer saccharinum and A. sac2 is Acer saccharum. For (a), thermal units are calculated as 22°C times the number of days in the experimental
conditions. For (b), the thermal units are 22°C times the number of days in the experimental conditions plus 13, the thermal units experienced outside before
collection. For (c), thermal units are based on weather records collected at the Blue Hills Meteorological Observatory (data from ncdc.noaa.gov). It should be
noted that the y-axis ranges from 0 to 800 for (a) and (b), and from 0 to 300 for (c). Thermal units were calculated in order to put the field and experimental data
on comparable scales; however, it is important to note that thermal units were calculated differently for the field and laboratory study, and the plants in each
study probablyexperienced thermal units differently (e.g. becauseday and night temperatures did not differ in the experimental set-up,plants in the laboratory
probably experienced more thermal units than they could developmentally use). Functional group: blue diamonds, invasive shrub; red squares, native shrub;
green triangles, native tree.

� 2013 The Authors
New Phytologist � 2013 New Phytologist Trust

New Phytologist (2013)
www.newphytologist.com

New
Phytologist Research 7



In particular, we found that 23 woody species in Concord, MA
are now leafing out an average of 18 d earlier than they did in Henry
David Thoreau’s time (c. 1850s) and that these changes are related
to warmer spring temperatures. On average, these species leafed out
5 d earlier for each 1°C warming in spring (March–April)
temperatures. This response is very similar to the sensitivity of
�4.6 d°C�1 (annual temperature) found in a meta-analysis of
temperate plants around the world (Wolkovich et al., 2012). It
appears that the temperature sensitivity of leaf-out phenology is
somewhat greater than the sensitivity of flowering phenology in
Concord, MA, which was �3.3 d °C�1 (Miller-Rushing &
Primack, 2008). Similar to flowering phenology, however, we
found that species that leaf early in the season show a stronger
response to temperature than those that leaf later in the growing
season. In the future, this difference in responsiveness may
contribute to a larger difference in leaf-out timing between early
and late leafing species.

Interspecific variation and implications for future forests

The speed of leaf out in the laboratory seemed to depend on two
factors: whether the chilling requirement had been met before
clipping; and the time needed for the plant to put out leaves in
response to warm conditions. In both years of the winter chilling
experiment, invasive shrubs showed the weakest chilling require-
ments, whereas native trees showed the strongest chilling
requirements, with native shrubs intermediate (Fig. 3). The time
needed for plants to put out leaves after they had met their
chilling requirements in March varied, with invasive shrubs
leafing out fastest and native shrubs and trees taking longer. In
both of these capacities – chilling requirements and speed of
putting out leaves once the chilling requirement is met (calculated
as both days and heat sums) – invasive shrubs, which already tend
to leaf out early in the season relative to natives, are poised to
respond to warming temperatures and shortening of the chilling
season in the wild faster than native shrubs and trees. They are
well positioned to take advantage of newly emerging temporal
niches early in the growing season before native canopy trees leaf
out (Richardson & Rejmanek, 2011; Fridley, 2012). Trees and
shrubs from more southern locations, which might have lower
chilling requirements, may also be able to enter this new early-
season niche.

The role of chilling

Although the results of our experiments revealed that most of the
species examined have winter chilling requirements which, if not
met, can delay the timing of leaf out, our field observations suggest
that chilling (or lack thereof) has not yet affected leaf-out dates in
wild-growing plants in Concord, MA. The lack of effect in the field
is evidence that, in the current climate, all or most of these species
are still meeting their chilling requirement in Massachusetts, and
the effects of winter chilling will only be seen with additional winter
warming. Even during the record warm winter of 2011–2012,
when there were only half as many chilling days as 2013, plants
apparently still received enough chilling to respond to the warm

Fig. 3 Proportion of total species in each functional group with minimal,
mild, moderate and strong winter chilling requirements. Functional groups
are invasive shrubs (n = 10), native shrubs (n = 20) and native trees (n = 16).

Table 3 Meandaystoleafoutin Marchand2013firstleaf-outdate(FLD)for
each functional group

Functional group Days to leaf out (� SD) 2013 FLD (� SD)

Invasive shrubs 10.7 � 3.2 (b) April 15 � 6.4 d (a)
Native shrubs 16.9 � 8.2 (b) April 27 � 8.8 d (b)
Native trees 26.8 � 12.3 (a) April 28 � 8.2 d (b)

Values given are the number of days taken by species in each functional
group to leaf out in the laboratory, and the mean first leaf date in the field in
2013. Functional groups include invasive shrubs (n = 10), native shrubs
(n = 20) and native trees (n = 16). Within columns, groups that share a letter
(a or b) are not significantly different (P > 0.05) as determined by Tukey’s
honestly significant difference test.

Minimal Mild Moderate Strong

05/14

05/10

05/06

05/02

04/28

04/24

04/20

04/16

04/12

04/08

M
ea

n 
FL

D
 (m

on
th

/d
ay

/y
ea

r)

Chilling requirement

Fig. 4 First leaf-out dates (FLDs) of species in each chilling requirement
category. Chilling requirements include minimal (11), mild (15), moderate
(17)andstrong(7).Central linesaremeanFLDs,boxboundariesarequartiles
and boundaries beyond boxes indicate the range of FLDs. The outlier in the
minimal chilling category is the invasive vine bittersweet (Celastrus
orbiculatus).

New Phytologist (2013) � 2013 The Authors
New Phytologist � 2013 New Phytologist Trustwww.newphytologist.com

Research

New
Phytologist8



spring and leaf out early. Abundant early season warmth may have
compensated for species that did not meet their full chilling
requirements. Observations in the southeastern USA have shown
delays in leaf out linked to insufficient winter chilling (Zhang et al.,
2007). Our evidence suggests eventual delays will probably occur
further north with more warming.

Photoperiod had little or no effect on leaf-out phenology.
American beech was the only species in which we detected a weak
photoperiod requirement on leafing out during the pilot study, as
shown in earlier studies (Heide, 1993b). Laube et al. (2013) also
found that photoperiod has a minor effect on leaf out compared
with chilling time. Our results suggest that photoperiod require-
ments will play only a minor role in plant leaf-out responses to a
warming climate in southern New England; however; further
experimental work is needed to determine the precise relationship
between photoperiod and leaf out in North America.

Methodological considerations

There may be some concern that the method of cutting twigs from
plants growing outside and exposing them to warm laboratory
conditions does not result in realistic experimental conditions.
However, this method is increasingly being used to investigate
leafing out requirements (Heide, 1993a; Ghelardini et al., 2010;
Laube et al., 2013). The best indication that cut twigs in the
laboratory serve as an accurate proxy for trees leafing out in the field
is that the order of leaf out following March collection was strongly
correlatedwith the order ofleafout inthe springof 2013. The order
of leaf out in January was also strongly correlated with the order of
leaf out in the field, indicating that the order of leaf out remains
largelythesameregardlessofthelengthofchilling.Itisalsotruethat
not all twigs and species responded well to the twig cutting process.
For example, red maple (Acer rubrum), grape and the two Lonicera
species used in the main study did initially leaf out, but the leaves
either wilted or did not develop further. However, the great
majorityoftwigsinourexperimentleafedoutnormallyandshowed
no signs of ill health. Researchers using this method need to take
precautions to ensure the health and responsiveness of the twigs.

Future work

The combination of using observational and historical data from
the field and experimental data represents an unusual, but useful
way to examine some of the complex factors affecting the leaf-out
response of plants to temperature, how the timing of leaf out has
changed over time and the physiological factors behind these
responses. Our study also demonstrates that the experimental
method of using cut twigs can be employed to effectively examine
the leaf-out requirements of entire communities of woody species.
We suggest that comparable surveys of plant communities from
other locations and surveys of widespread species across latitudinal
gradients could provide further ecological and evolutionary
insights into the leaf-out process during this period of climate
change.

Acknowledgements

We would like to thank Abe Miller-Rushing, Pamela Templer,
Michael Dosmann, Jenica Allen, Ines Ibanez, John Silander, Libby
Ellwood and three anonymous reviewers for comments on the
paper. Thanks to Chris Perriello, Jean Polgar, Emma Polgar, Libby
Ellwood, Caitlin McDonough MacKenzie, Sang Don Lee,
Julianne Richards, Viet Le, Corbin Kuntze, Nicole Chabaneix,
Michelle McInnis, Rika Fuchino, Peter Everill, Jessica Susser and
Meg Boeni for help with laboratory work and fieldwork. We also
thank the Arnold Arboretum of Harvard University and other
landowners for allowing us to collect twigs. Funding for the project
came from National Science Foundation (NSF) grant DEB-
0842749.

References

Archetti M, Richardson AD, O‘Keefe J, Delpierre N. 2013. Predicting climate

change impacts on the amount and duration of autumn colors in a New England

forest. PloS ONE 8: e57373.
Ault TR, Henebry GM, de Beurs KM, Schwartz MD, Betancourt JL, Moore D.

2013. The false spring of 2012, earliest in North American Record. Eos
Transactions American Geophysical Union 94: 181–182.

Basler D, Korner C. 2012. Photoperiod sensitivity of bud burst in 14 temperate

forest tree species. Agricultural and Forest Meteorology 165: 73–81.
Cleland EE, Chuine I, Menzel A, Mooney HA, Schwartz MD. 2007. Shifting plant

phenology in response to global change. Trends in Ecology & Evolution 22:
357–365.

Cook BI, Wolkovich EM, Parmesan C. 2012. Divergent responses to spring and

winter warming drive community level flowering trends. Proceedings of the
National Academy of Sciences, USA 109: 9000–9005.

Ellwood E, Temple SA, Primack RB, Bradley NL, Davis CC. 2013.

Record-breaking early flowering in the eastern United States. PLoS ONE 8:
e53788.

Falusi M, Calamassi R. 1990. Bud dormancy in beech (Fagus sylvatica L.). Effect of
chilling and photoperiod on dormancy release of beech seedlings. Tree Physiology
6: 429–438.

Faust M, Erez A, Rowland LJ, Wang SY, Norman HA. 1997. Bud dormancy in

perennial fruit trees: Physiological basis for dormancy induction, maintenance,

and release. Hortscience 32: 623–629.
Fridley JD. 2012. Extended leaf phenology and the autumn niche in deciduous

forest invasions. Nature 485: 359–362.
Fu YSH, Campioli M, Van Oijen M, Deckmyn G, Janssens IA. 2012. Bayesian

comparison of six different temperature-based budburst models for four

temperate tree species. Ecological Modelling 230: 92–100.
Ghelardini L, Santini A, Black-Samuelsson S, Myking T, Falusi M. 2010. Bud

dormancy release in elm (Ulmus spp.) clones – a case study of photoperiod and
temperature responses. Tree Physiology 30: 264–274.

Gu L, Hanson PJ, Mac Post W, Kaiser DP, Yang B, Nemani R, Pallardy SG,

Meyers T. 2008. The 2007 eastern US spring freezes: increased cold damage in a

warming world? BioScience 58: 253–262.
Heide OM. 1993a. Daylength and thermal time responses of budburst during

dormancy release in some northern deciduous trees. Physiologia Plantarum 88:
531–540.

Heide OM. 1993b. Dormancy release in beech buds (Fagus sylvatica) requires both
chilling and long days. Physiologia Plantarum 89: 187–191.

Hulme PE. 2011. Contrasting impacts of climate-driven flowering phenology on

changes in alien and native plant species distributions. New Phytologist 189:
272–281.

Korner C, Basler D. 2010. Phenology under global warming. Science 327:
1461–1462.

� 2013 The Authors
New Phytologist � 2013 New Phytologist Trust

New Phytologist (2013)
www.newphytologist.com

New
Phytologist Research 9



Laube J, Sparks TH, Estrella N, Hofler J, Ankerst DP, Menzel A. 2013. Chilling

outweighs photoperiod in preventing precocious spring development. Global
Change Biology 20: 170–182.

Lechowicz MJ. 1984. Why do temperate deciduous trees leaf out at different times –
adaptations and ecology of forest communities. American Naturalist 124:
821–842.

McCreary DD, Lavender DP, Hermann RK. 1990. Predicted global warming

and Douglas-fir chilling requirements. Annales des Sciences Forestieres 47:
325–330.

Miller-Rushing AJ, Primack RB. 2008. Global warming and flowering times in

Thoreau’s concord: a community perspective. Ecology 89: 332–341.
Morin X, Lechowicz MJ, Augspurger C, O’ Keefe J, Viner D, Chuine I. 2009. Leaf

phenology in 22 North American tree species during the 21st century. Global
Change Biology 15: 961–975.

Murray MB, Cannell MGR, Smith RI. 1989. Date of budburst of 15 tree species in

Britain following climatic warming. Journal of Applied Ecology 26: 693–700.
Polgar CA, Primack RB. 2011. Leaf-out phenology of temperate woody plants:

from trees to ecosystems. New Phytologist 191: 926–941.
Primack RB, Miller-Rushing AJ. 2012. Uncovering, collecting, and analyzing

records to investigate the ecological impacts of climate change: a template from

Thoreau’s Concord. BioScience 62: 170–181.
Primack RB, Miller-Rushing AJ, Dharaneeswaran K. 2009. Changes in the flora of

Thoreau’s Concord. Biological Conservation 142: 500–508.
Richardson AD, Rejmanek M. 2011. Trees and shrubs as invasive alien species – a
global review. Diversity and Distributions 17: 788–809.

Vitasse Y, Basler D. 2013. What role for photoperiod in the bud burst phenology of

European beech. European Journal of Forest Research 132: 1–8.
Vitasse Y, Delzon S, Dufrene E, Pontailler JY, Louvet JM, Kremer A, Michalet R.

2009. Leaf phenology sensitivity to temperature in European trees: do

within-species populations exhibit similar responses? Agricultural and Forest
Meteorology 149: 735–744.

Willis CG, Ruhfel B, Primack RB, Miller-Rushing AJ, Davis CC. 2008.

Phylogenetic patterns of species loss in Thoreau’s woods are driven by

climate change. Proceedings of the National Academy of Sciences, USA 105: 17029–
17033.

Willis CG, Ruhfel BR, Primack RB, Miller-Rushing AJ, Losos JB, Davis CC.

2010. Favorable climate change response explains non-native species’ success in

Thoreau’s woods. PLoS ONE 5: e8878.
Wolkovich EM, Cook BI, Allen JM, Crimmins TM, Betancourt JL, Travers SE,

Pau S, Regetz J, Davies TJ, Kraft NJB et al. 2012. Warming experiments
underpredict plant phenological responses to climate change. Nature 485:
494–497.

Xu C, Griffin KL, Schuster WSF. 2007. Leaf phenology and seasonal variation of

photosynthesis of invasive Berberis thunbergii (Japanese barberry) and two
co-occurring native understory shrubs in a northeastern United States deciduous

forest. Oecologia 154: 11–21.
Yu H, Luedeling E, Xu J. 2010. Winter and spring warming result in delayed spring

phenology on the Tibetan Plateau. Proceedings of the National Academy of Sciences,
USA 107: 22151–22156.

Zhang XY, Tarpley D, Sullivan JT. 2007. Diverse responses of vegetation

phenology to a warming climate. Geophysical Research Letters 34: L19405.

Supporting Information

Additional supporting information may be found in the online
version of this article.

Fig. S1 Mean combined spring (March and April) temperatures in
Massachusetts for the years 1852–2013.

Table S1 Collection dates for twigs in the 2011–2012 pilot study
and the 2013 main study, together with chilling units and thermal
units that had occurred up to the day of collection, and the day
length

Table S2 Days to first leaf out for study species for each collection
period under the natural day treatment and chilling categories for
the first year pilot laboratory study

Table S3 Results of simple linear regressions measuring the
response of leaf-out timing to mean March and April temperature
for the field study in Concord, MA, USA

Please note: Wiley Blackwell are not responsible for the content or
functionality of any supporting information supplied by the
authors. Any queries (other than missing material) should be
directed to the New Phytologist Central Office.

New Phytologist (2013) � 2013 The Authors
New Phytologist � 2013 New Phytologist Trustwww.newphytologist.com

Research

New
Phytologist10