Content uploaded by Carl A Roland
Author content
All content in this area was uploaded by Carl A Roland on Dec 03, 2018
Content may be subject to copyright.
PRIMARY RESEARCH ARTICLE
Stand basal area and solar radiation amplify white spruce
climate sensitivity in interior Alaska: Evidence from carbon
isotopes and tree rings
Elizabeth Fleur Nicklen
1,2
|
Carl A. Roland
1,3
|
Adam Z. Csank
4
|
Martin Wilmking
5
|
Roger W. Ruess
2
|
Laurel Ann Muldoon
6
1
Central Alaska Network, National Park
Service, Fairbanks, Alaska
2
Department of Biology and Wildlife,
Institute of Arctic Biology, University of
Alaska Fairbanks, Fairbanks, Alaska
3
Denali National Park and Preserve,
Fairbanks, Alaska
4
Department of Geography, University of
Nevada Reno, Reno, Nevada
5
Institute of Botany and Landscape
Ecology, Ernst‐Moritz‐Arndt University
Greifswald, Greifswald, Germany
6
Department of Environmental Geography,
Nipissing University, North Bay, Ontario,
Canada
Correspondence
Elizabeth Fleur Nicklen, Central Alaska
Network, National Park Service, Fairbanks,
AK.
Email: elizabeth_nicklen@nps.gov
Funding information
National Park Service
Abstract
The negative growth response of North American boreal forest trees to warm sum-
mers is well documented and the constraint of competition on tree growth widely
reported, but the potential interaction between climate and competition in the bor-
eal forest is not well studied. Because competition may amplify or mute tree cli-
mate‐growth responses, understanding the role current forest structure plays in tree
growth responses to climate is critical in assessing and managing future forest pro-
ductivity in a warming climate. Using white spruce tree ring and carbon isotope data
from a long‐term vegetation monitoring program in Denali National Park and Pre-
serve, we investigated the hypotheses that (a) competition and site moisture charac-
teristics mediate white spruce radial growth response to climate and (b) moisture
limitation is the mechanism for reduced growth. We further examined the impact of
large reproductive events (mast years) on white spruce radial growth and stomatal
regulation. We found that competition and site moisture characteristics mediated
white spruce climate‐growth response. The negative radial growth response to
warm and dry early‐to mid‐summer and dry late summer conditions intensified in
high competition stands and in areas receiving high potential solar radiation. Dis-
crimination against
13
C was reduced in warm, dry summers and further diminished
on south‐facing hillslopes and in high competition stands, but was unaffected by cli-
mate in open floodplain stands, supporting the hypothesis that competition for
moisture limits growth. Finally, during mast years, we found a shift in current year's
carbon resources from radial growth to reproduction, reduced
13
C discrimination,
and increased intrinsic water‐use efficiency. Our findings highlight the importance of
temporally variable and confounded factors, such as forest structure and climate, on
the observed climate‐growth response of white spruce. Thus, white spruce growth
trends and productivity in a warming climate will likely depend on landscape posi-
tion and current forest structure.
KEYWORDS
boreal forest, carbon isotopes, climate change, climate‐growth response, competition,
dendroecology, Picea glauca
----------------------------------------------------------------------------------------------------------------------------------------------------------------------
Published 2018. This article is a U.S. Government work and is in the public domain in the USA
Received: 1 June 2018
|
Accepted: 18 October 2018
DOI: 10.1111/gcb.14511
Glob Change Biol. 2018;1–16. wileyonlinelibrary.com/journal/gcb
|
1
1
|
INTRODUCTION
Forest structure and demography are changing with a warming cli-
mate. Widespread increases in tree mortality have occurred in the
western United States (van Mantgem et al., 2009) and Canada (Luo
& Chen, 2015; Peng et al., 2011; Zhang, Huang, & He, 2015), reduc-
ing stand basal area and shifting stand age. Changing disturbance
regimes associated with climate warming further alter forest struc-
ture and demography (Barrett, McGuire, Hoy, & Kasischke, 2011;
Johnstone, Hollingsworth, Chapin, & Mack, 2010). Forest structure
and demography are a product of, and feed back to tree growth,
death, and recruitment rates. In particular, competition is known to
have large, effects on tree growth, recruitment, and mortality
(Aakala, Fraver, D'Amato, & Palik, 2013; Alam et al., 2017; Coomes
& Allen, 2007; Cortini, Comeau, & Bokalo, 2012; Fernández‐de‐Uña,
Cañellas, & Gea‐Izquierdo, 2015; Trugman, Medvigy, Anderegg, &
Pacala, 2017; Zhang et al., 2015). Nonetheless, competition is fre-
quently not considered in tree climate‐growth analyses, though it
has been getting more attention recently in a variety of forest types
around the world (Alam et al., 2017; Cortini et al., 2012; Fernández‐
de‐Uña et al., 2015; Fernández‐de‐Uña, McDowell, Cañellas, & Gea‐
Izquierdo, 2016; Piutti & Cescatti, 1997; Ruiz‐Benito et al., 2014)
including in the boreal forest in interior (Trugman et al., 2017) and
southwest Alaska (Wright, Sherriff, Miller, & Wilson, 2018). Studies
of productivity in mature stands in boreal forests have found nega-
tive growth responses to warm summer temperatures (Angert et al.,
2005; Barber, Juday, & Finney, 2000; Beck et al., 2011; Bunn &
Goetz, 2006; Girardin et al., 2016; Huang et al., 2010; Jiang et al.,
2016; Juday & Alix, 2012; Juday, Alix, & Grant, 2015; Lloyd, Duffy,
& Mann, 2013; Sullivan, Pattison, Brownlee, Cahoon, & Hollings-
worth, 2017; Walker & Johnstone, 2014; Zhang et al., 2015), though
these climate‐growth responses can be variable within (Wilmking,
Juday, Barber, & Zald, 2004; Wilmking, Juday, Terwilliger, & Barber,
2006) or across site physical (Nicklen, Roland, Ruess, Schmidt, &
Lloyd, 2016), vegetative (Bunn & Goetz, 2006), and regional (Hell-
mann et al., 2016; Lloyd & Bunn, 2007) conditions.
Though the negative growth response of non‐treeline boreal
spruce forests to warm growing seasons is well documented (Barber
et al., 2000; Juday & Alix, 2012) and the strong negative effect of
competition on tree growth is known, the likely interactive effect of
competition on climate‐growth responses is poorly studied. The
potential interactive effect of competition and climate may confound
our understanding of climate‐growth responses (e.g., negative growth
responses attributed to climate may be partially related to stand
competition levels), but if quantified, may elucidate unexplained vari-
ability in climate‐growth response within and across boreal forest
stands. The potential interactive effect between competition and cli-
mate may result in unexpectedly amplified or muted growth
responses to climate affecting the future structure and demography
of the boreal forest. This gap in our understanding of boreal forest
climate‐growth relationships is of particular concern for two reasons.
First, the climate is warming at a faster rate in high latitude forests
compared to lower latitude forests (Hinzman et al., 2005; IPCC,
2014) and the unknown interaction of this warming with tree
growth presents greater uncertainly about future boreal forest struc-
ture and function. Second, the boreal forest represents nearly a third
of the world's forests (Kuusela, 1990) and 22% of the carbon storage
on Earth's land surface (IPCC, 2014). Thus, the growth response of
trees in the boreal forest to climate change will greatly affect the
future footprint of the boreal biome, carbon dynamics (Chapin et al.,
2009; Cox, Betts, Jones, Spall, & Totterdell, 2000; Koven, 2013; Sha-
ver, Billings, Chapin, Giblin, & Nadelhoffer, 1992), and albedo (Betts
& Ball, 1997; Bonan, 2008; Euskirchen, McGuire, Rupp, Chapin, &
Walsh, 2009) and influence habitat for flora and fauna over large
areas.
White spruce (Picea glauca) is vulnerable to decreased water
availability (Barber et al., 2000; McGuire et al., 2010; Yarie & Van
Cleve, 2010; Yarie, Cleve, & Schlentner, 1990). Thus, white spruce in
high basal area stands, where competition for water may be
increased, could suffer exacerbated drought stress during warm, dry
growing seasons relative to spruce in more open stands with less
competition. Indeed, white spruce growth response to climate in
western Canada was reduced when growing in stands with high
aspen basal area (Cortini et al., 2012). Similarly, the correlation
between growth and temperature for European beech changed from
positive to negative as competition increased (Piutti & Cescatti,
1997). In central Canada, climate enhanced conspecific competition
increased mortality in Pinus banksiana and Populus tremuloides (Luo &
Chen, 2015). There is some evidence for similar climate–competition
interactions for white spruce in interior Alaska: Growth was nega-
tively correlated with soil moisture deficit in unthinned, but not
thinned stands (Yarie et al., 1990) and more positively associated
with precipitation in mature forests than in open treeline stands
(Ohse, Jansen, & Wilmking, 2012).
Stable carbon isotopic ratios in tree rings record the balance
between stomatal conductance and photosynthetic rate (Farquhar,
O'Leary, & Berry, 1982; Francey & Farquhar, 1982) providing annu-
ally resolved information about physiological responses to environ-
mental conditions (McCarroll & Loader, 2004). Thus, environmental
conditions that influence stomatal conductance (soil water availabil-
ity, vapor pressure deficit) and photosynthesis (air temperature,
nutrient availability, irradiance) will also be recorded in the isotopic
ratios of tree rings. During photosynthesis, ribulose‐1,5‐biphosphate
carboxylase/oxygenase discriminates against the heavier
13
C in favor
of
12
C contributing to a lower
13
Cto
12
C ratio (δ
13
C) in the leaves
and wood of trees than in the atmosphere (Farquhar et al., 1982).
When drought stressed, trees limit water loss from transpiration by
closing stomata, which also limits the atmospheric CO
2
available for
photosynthesis, forcing increased assimilation of
13
C during carboxy-
lation and leading to greater intrinsic water‐use efficiency (iWUE).
Carbohydrates produced under drought stress conditions have a
higher δ
13
C value reflecting less
13
C discrimination (Δ
13
C) and higher
iWUE. This drought stress signature is preserved within the annual
growth rings of trees. Arctic treeline white spruce stomatal closure
has been observed in response to high vapor pressure deficit and is
exacerbated by limited soil water availability (Sullivan &
2
|
NICKLEN ET AL.
Sveinbjörnsson, 2011). In Alaska and western Canada, white spruce
carbon isotopic ratios are also correlated with temperature (Barber
et al., 2000; Csank, Miller, Sherriff, Berg, & Welker, 2016; Holzkäm-
per, Tillman, Kuhry, & Esper, 2012; Porter, Pisaric, Kokelj, &
Edwards, 2009) and relative humidity (Porter et al., 2009).
In this study, we investigate potential interactive effects of com-
petition as well as site moisture characteristics with climate on white
spruce radial growth and stomatal regulation. Using tree ring and
carbon isotope data from a long‐term vegetation monitoring program
in Denali National Park and Preserve (see Roland, Schmidt, & Nick-
len, 2013), we address the hypotheses that (a) white spruce growth
response to climate can be mediated by stand competition and site
moisture characteristics and (b) moisture limitation is the mechanism
for reduced growth in warm, dry years. If competition does mediate
climate‐growth responses, we expect radial growth will have a less
positive or more negative response to warm, dry growing seasons in
high competition stands compared with radial growth in open stands.
If competition for water (as opposed to nutrients or light) is driving
the reduction in radial growth, we expect reduced ring growth and
decreased
13
C discrimination (and increased iWUE) during warm, dry
growing seasons in dry vs. moist sites and in high vs. low competi-
tion stands. If competition for nutrients or light limits growth rather
than competition for water, photosynthesis may be reduced relative
to stomatal conductance, and we expect reduced radial growth in
high BA sites in concert with either no change in or increased Δ
13
C
(and decreased iWUE).
2
|
MATERIALS AND METHODS
2.1
|
Study area
The 1.28 million ha study area is situated in south‐central interior
Alaska within Denali National Park and Preserve (DNPP), almost
entirely on the north side of the Alaska Range within DNPP (center
near 63°41′N, 150°25′W; see Roland et al., 2013). The area includes
steep Alaska Range hillslopes >2,400 m elevation, foothill ranges,
and extensive lowland basins. Permafrost is continuous to discontin-
uous in the lowland basins and discontinuous to sporadic in the
Alaska Range (Clark & Duffy, 2006). The study area experiences a
continental climate with very cold, dry winters and short, warm sum-
mers. Temperature and precipitation are variable across the study
area, which spans two climate regions (Bieniek et al., 2012). Mean
January temperatures are ~−22°C in the lowland basins in the NW
end of the park and ~−12°C in the Alaska Range. Mean July temper-
atures are ~16°C in the basins and ~8°C in the Alaska Range (1971–
2000 monthly PRISM averages; Daly, 2009). Annual precipitation
sums range from less than 400 mm in the lowlands to over
1,000 mm in the mountains (1971–2000 monthly PRISM averages;
Daly, 2009), with the majority falling in June through August (Sou-
sanes, 2008). Fire return intervals in the most fire‐prone area of the
park, the NW lowland basin, are around 200–300 years (Kasischke,
Williams, & Barry, 2002). Fires are much less frequent at higher ele-
vations.
Six tree species occur in our study area: white spruce (Picea
glauca), black spruce (Picea mariana), Alaska birch (Betula neoalas-
kana), trembling aspen (Populus tremuloides), balsam poplar (Populus
balsamifera), and tamarack (Larix laricina; Roland et al., 2013). Of
these, white and black spruce are by far the most frequent and
abundant trees in the study area. White spruce occupies 32% of
plots in the study area with a mean basal area (BA) for occupied
plots of 6.0 m
2
/ha and occurs in a wide range of topographic posi-
tions, including being the most common treeline species in DNPP
(Roland et al., 2013). The highest BA white spruce stands occur on
warm, well‐drained south‐exposed mid‐elevation slopes, in per-
mafrost‐free terrain. Black spruce occurs in 27% of plots with a
mean BA of 1.0 m
2
/ha for occupied plots and is generally restricted
to the lowland basins and hills in DNPP primarily in areas affected
by permafrost (Roland et al., 2013). The remaining four tree species
each occur in 13% or fewer plots in the study area and are relative
habitat specialists (Roland et al., 2013). Alaska birch and trembling
aspen are strongly associated with burned areas, where aspen occurs
in the warmest and driest sites. Balsam poplar is primarily found
along rivers and in gravelly soils, and tamarack is found within a sub-
set of black spruce habitat, generally in wet areas.
2.2
|
Study design
As a part of the National Park Service's inventory and monitoring
program, a systematic sampling grid was established across the study
area (Roland et al., 2013; Roland, Oakley, Debevec, & Loomis, 2004).
Grid spacing was 20‐km, but with 10‐km spacing within 6 km of the
one park road (Figure 1). At each grid, intersection was a “mini‐grid”
consisting of five rows of five plots spaced 500 m apart. Each of the
25 points within a “mini‐grid”entailed a circular 16 m diameter plot
and an outer meta‐plot extending another 10 m in radius.
At each plot, we measured a suite of topographic, edaphic, and
vegetative variables (details in Roland et al., 2004, 2013). In addition
to slope angle, elevation, and aspect, we recorded the diameter of
each tree species at 1.37 m above ground level (DBH) within the
plot and calculated plot basal area (BA; m
2
/ha). We used point inter-
cept transects to estimate percent tall shrub cover. At four cardinal
directions within the meta‐plot, we dug soil pits where, in addition
to collecting soil for physical and chemical analysis, we measured the
depth of the living mat and the soil organic layer (SOL) and recorded
soil temperature at 10 cm below soil surface. We classified plots as
either having growing season shallow frozen soil (GsSFS) or not. A
plot was considered to have frozen soil during the growing season if
the average of 16 soil depths measured within each plot using a
130‐cm soil probe was less than 50 cm and we found ice in at least
one of the four soil pits or if the four soil temperatures taken at the
plot were all <1°C. Otherwise, the plot was classified as free of near
surface frozen soil during the growing season. We obtained potential
solar radiation receipts for each of our plot locations using the Solar
Analyst tool in ArcGIS 10.0 (Dubayah & Rich, 1995), which incorpo-
rates slope angle, aspect, latitude, sun angle, and surrounding land-
scape (based on the United States Geological Survey 60‐m digital
NICKLEN ET AL.
|
3
elevation model from the National Elevation Dataset) into monthly
radiation value estimates (Rich et al., 1994). We summed the
monthly values into one potential solar radiation estimate per plot.
We assigned each plot one of five lithology categories (bedrock, allu-
vium, drift, Nenana gravels, and eolian) based on the Denali Soil
Map (Clark & Duffy, 2006). Sampling occurred between 2003 and
2010 during the months of June, July, and August.
2.3
|
Tree sampling
Of the 43 mini‐grids and 1,107 plots sampled, we cored and cross‐
dated 357 white spruce trees 10,347 growth rings) from a total of
26 mini‐grids and 160 plots. We cored the largest tree within each
of the four quadrants of the meta‐plot, though often there were
fewer than four trees to core or the trees were too small to core
(<5 cm at 1.37 m from ground surface; see Supporting Information
Figure S2 for size bias in cored trees). Many of the plots contained
no trees as they were above treeline, too wet or too recently dis-
turbed. We extracted penetrating cores whenever possible and
cored as low to the ground as possible while avoiding bole
deformities. Dried, mounted, and sanded cores were measured to
the nearest 0.001 mm and cross‐dated (see Nicklen et al., 2016).
Dating was validated with COFECHA and the Dendrochronology
Program Library in R (dplR, Bunn, 2008; Bunn & Korpela, 2016). We
averaged replicate rings widths from individual trees by year. We
used ring widths and tree radius to calculate basal area increment
(BAI), an estimate of the area of wood produced by each tree in
each year of growth (Nicklen et al., 2016). Because we were inter-
ested in the effect of biotic variables on growth, and biotic variables
(moss depth, stand basal area) change over time, we limited our ring
width sample to within 30 years of the sampling date as a balance
between sample size and changing site conditions. The earliest years
considered in our sample ranged from 1974 to 1981.
We selected six trees from each of two mini‐grids for additional
carbon isotope analysis (Figure 1). To target our interest in competi-
tion and drought stress, we selected trees and mini‐grids based on
(a) capturing a large gradient in tree BA across the plots (competition
proxy), (b) one mini‐grid was to be located in a floodplain and the
other on a south‐facing hillslope, and (c) other variables such as tree
age and size, elevation, and general mini‐grid location were to be
FIGURE 1 Map showing the location of sampled white spruce from minigrids in the Denali National Park and Preserve study area
4
|
NICKLEN ET AL.
kept as similar as possible, such that BA and site moisture were the
focal covariates (Figure 1; Supporting Information Table S1). All
selected trees were cored in 2009 or 2010, and Δ
13
C analysis was
conducted on each ring with sufficient wood going back to 1980.
Each ring was sampled using a Foredom Flexshaft Drill (Foredom
Electric Company, Bethel, CT, USA) fixed in place under a micro-
scope. Samples were subsequently processed to ɑ‐cellulose using
the modified version of the Brendel, Iannetta, and Stewart (2000)
method recommended by Anchukaitis et al. (2008). Samples were
weighed into tin capsules and analyzed for δ
13
C using a Delta V
Advantage with EA in the Organic Biogeochemistry Laboratory at
the University of Notre Dame. The combustion reactor was run at
1,000°C, reduction reactor at 650°C, with the column at 65°C. Each
run took 450 s with three reference gas peaks run at the beginning
and end of the run (N2 at the beginning and CO
2
at the end). The
CO
2
ran with a 75% dilution. Analytical precision was 0.2‰. Carbon
isotope discrimination (Δ
13
C) was calculated as:
Δ13C¼δ13 Caδ13Ctree
1þδ13Ctree =1000
Annual estimates of δ
13
C
a
are the δ
13
C value of atmospheric
CO
2
obtained from flask data collected at Point Barrow station for
the years 1982 to 2009 (Keeling et al., 2001) and from McCarroll
and Loader (2004) for 1980 and 1981. δ
13
C
tree
is the value of δ
13
C
measured from tree‐ring cellulose for a given year. We calculated
intrinsic water‐use efficiency (iWUE) using two equations, from Far-
quhar et al. (1982):
Δ13C¼aþðbaÞ Ci
Ca
where ais the fractionation of
13
CO
2
relative to
12
CO
2
during diffu-
sion (4.4‰), bis the biochemical fractionation during photosynthesis
(27‰), and c
i
/c
a
is the ratio of CO
2
inside the leaf to CO
2
in the
atmosphere. Solving for c
i
and using the ratio of diffusivities of
water vapor and c
a
, we calculate iWUE (Farquhar, Hubick, Condon,
& Richards, 1989):
iWUE ¼A
Gs
¼ðcaciÞð1=1:6Þ
2.4
|
Model covariates
We focused on site and tree variables expected to represent or
influence competition and moisture availability, but additionally
TABLE 1 Characteristics of covariates used in models of white spruce radial growth for the entire DNPP sample (for characteristics of
subset trees used in δ
13
C analysis, see Supporting Information Table S1)
Variable Mean SD Min Max
Competition factors Climate interaction tested?
Conifer basal area (m
2
/ha) 11.0 11.9 0 66 Yes
Broadleaf basal area (m
2
/ha) 3.2 6.9 0 35 No
Shrub cover above 1.5 m (%) 17.3 22.8 0 88 Yes
Drought stress/site moisture factors
Slope angle (degrees) 9.0 7.8 1 36 Yes
Potential solar radiation (KWH/m
2a
) 680.4 66.0 486 839 Yes
Live mat depth (cm) 4.2 3.3 0 21 Yes
Soil organic mat (SOL; cm) 14.0 7.6 1 30 Yes
Tree size (DBH, cm) 22.7 9.7 5 54 Yes
Factors known to influence P. glauca growth in DNPP
Lithology (five classes) NA NA NA NA No
Growing season shallow frozen soil (GsSFS; binary) NA NA NA NA No
Spruce mast year (binary) NA NA NA NA No
Minimum age at time of coring 116 57.4 20 343 No
Climate factors Prev. and/or current year?
Precip/snow Oct–Apr (mm) 182.9 67.8 57 430 Current
VPD May
b
(hPa) 5.5 1.1 2.3 10.1 Current
Precip. May (mm) 26.2 12.3 6 64 Current
VPD Jun–Jul
b
(hPa) 6.6 1.4 2.3 11.8 Previous & current
Precip. Jun–Jul (mm) 152.6 58.7 51 419 Previous & current
VPD Aug
b
(hPa) 4.0 1.2 1.1 8.9 Previous
Precip. Aug (mm) 76.5 27.8 20 172 Previous
Notes. Whether the site or tree covariate's interactive effect with climate was tested is indicated as well as whether the climate variable used was for
the year current with and/or previous to the year of ring formation.
a
Solar radiation units are sum of kilowatt hours per square meter.
b
Quadratic terms evaluated in competing models.
NICKLEN ET AL.
|
5
included variables known to influence white spruce growth in DNPP
(Table 1). Conifer BA was our primary metric of competition.
Although BA is a measure of stand productivity, it has also been suc-
cessfully used as a measure of plot crowding as it integrates both
stand density and tree size (Martin & Ek, 1984; Trugman et al.,
2017). We focused on within conifer competition for two reasons.
First, competition within species or among congeners (e.g., Picea
glauca and Picea mariana) is expected to be more intense than inter-
specific competition as different species or functional groups differ
in resource use and phenology (Man & Lieffers, 1999). Second, there
was considerably more conifer basal area across our study than
broadleaf basal area (Table 1, Roland et al., 2013) making this data-
set unsuitable for testing white spruce–broadleaf competition. We
did include broadleaf basal area as a main effect to account for
those few areas where there was high broadleaf BA. We included
individual tree DBH both as a measure of tree relative dominance
and as a potential variable in drought stress susceptibility, as larger
trees can be more sensitive to drought than smaller trees (Bennett,
Mcdowell, Allen, & Anderson‐Teixeira, 2015). We assessed tall shrub
cover as a possible competition factor as tall shrubs have been
found to negatively affect white spruce growth (Cortini & Comeau,
2008). We evaluated plot slope angle and potential solar radiation
estimates as site variables that may influence site moisture balance
and tested depth of living mat and SOL as variables likely to influ-
ence soil moisture availability and climate‐growth responses (Droby-
shev, Simard, Bergeron, & Hofgaard, 2010; Gewehr, Drobyshev,
Berninger, & Bergeron, 2014). Finally, we considered several vari-
ables known to influence white spruce growth in DNPP as main
effects only: ring age (ring count), lithology type, and the presence
or absence of growing season shallow frozen soil (Nicklen et al.,
2016). We modeled ring age as a cubic function to account for
growth trends related to age (Nicklen et al., 2016). We also included
a binary variable to indicate mast seeding years in spruce as mast
years are strongly associated with reduced radial growth of that year
(Juday, Barber, Rupp, Zasada, & Wilmking, 2003). Mast data were
based on records from interior Alaska (Juday et al., 2003; Roland,
Schmidt, & Johnstone, 2014).
We considered the following climate variables: mean vapor pres-
sure deficit (VPD) from June–July and August of the previous grow-
ing season and from May and June–July of the current growing
season (Table 2). We also considered precipitation sums from the
same time periods, as well as from the winter season (October–April)
prior to ring formation. These data are downscaled and spatially
interpolated climate data provided by Scenarios Network for Alaska
and Arctic Planning (SNAP 2014; Retrieved in 2014 from https://
www.snap.uaf.edu/tools/data-downloads; temperature and vapor
pressure from CRU TS 3.1, precipitation from CRU TS3.1.01) for the
years 1974–2009. The climate data are estimates of historical
monthly climatic variables for any given locale in Alaska at 1‐km res-
olution. SNAP downscaled monthly climate data for Alaska to a finer
grid resolution using PRISM (Parameter–elevation Relationships on
Independent Slopes Model) which integrates location, elevation,
coastal proximity, topographic variables, vertical atmospheric layer,
and orographic effectiveness of the terrain (Daly et al., 2008). For
each plot and each year for which we had tree‐ring data within
1974–2009, we extracted the monthly mean vapor pressure (hPa),
temperature (°C), and precipitation sum (mm). We used the tempera-
ture data (Temp) to calculate mean monthly saturated vapor pressure
(SVP) in hectoPascals (hPa) from Murray (1967):
SVP ¼610:710ð7:5TempÞ=ð237:3þTempÞ=100
We then subtracted the mean monthly vapor pressure (VAP)
value to obtain mean monthly vapor pressure deficit (VPD = SVP −
VAP). Nonlinear climate‐growth relationships have been found for
white spruce (D'Arrigo et al., 2004; Lloyd et al., 2013; Nicklen et al.,
2016; Sullivan et al., 2017; Wilmking et al., 2004); thus, we included
nonlinear (quadratic) VPD terms (Table 2). As gridded precipitation
data can be imprecise (McAfee et al., 2014), we included only linear
TABLE 2 Standardized coefficient estimates for significant
relationships between log‐transformed white spruce BAI and
covariates in final selected model
Covariates and interactions Estimate
Lower
95%
limit
Upper
95%
limit
Intercept 5.799 5.740 5.858
Age proxy −0.126 −0.155 −0.096
Age proxy
2
−0.043 −0.060 −0.026
Age proxy
3
0.017 0.011 0.023
Broadleaf BA 0.047 0.007 0.086
DBH 0.224 0.188 0.260
Plot slope −0.037 −0.072 −0.001
Potential solar radiation −0.039 −0.067 −0.010
LiveMat_cm −0.058 −0.086 −0.031
Eolian lithology 0.332 0.185 0.478
Mast Yr −0.088 −0.101 −0.074
Precip/snow current Oct–Apr −0.030 −0.036 −0.024
VPD current May −0.015 −0.022 −0.009
VPD current May
2
0.016 0.012 0.019
VPD current Jun–Jul 0.017 0.008 0.027
VPD current Jun–Jul
2
−0.014 −0.018 −0.010
VPD previous Jun–Jul −0.071 −0.081 −0.061
VPD previous Jun–Jul
2
0.021 0.016 0.025
Precip previous Jun–Jul 0.010 0.004 0.016
Precip previous Aug 0.028 0.022 0.033
VPD cur. May ×solar rad. −0.013 −0.021 −0.006
VPD cur. May
2
×solar rad. −0.005 −0.009 −0.001
Precip. cur. May ×solar rad. 0.009 0.004 0.014
Precip. prev. Aug ×solar rad. −0.008 −0.013 −0.003
Conifer BA ×VPD prev. Jun–Jul −0.011 −0.019 −0.004
Conifer BA ×precip prev. Aug 0.010 0.005 0.016
Note. Covariates were scaled to have a mean of zero and a standard
deviation of 1; thus, effects are relative. The marginal and conditional R
2
for the model were 0.15 and 0.92, respectively. N= 357 trees and
10,347 growth rings.
6
|
NICKLEN ET AL.
precipitation–growth relationships to minimize detecting spurious
precipitation–growth relationships.
2.5
|
Modeling climate–competition effects on
white spruce radial growth (full DNPP tree sample)
We used linear mixed‐effects models to quantify the effect of com-
petition–climate and site moisture–climate interactions on the BAI
of white spruce. We log‐transformed BAI and standardized continu-
ous model covariates to have a mean of zero and a standard devia-
tion of one. Thus, model coefficients were standardized and
comparable. Because of the complexity of variables involved, we
used a three‐stage approach to model development. In the first
stage, we established a “base”model by comparing models with
the essential factors contributing to radial growth (ring age, tree
size, reproductive effort, climate variables, and competition effects)
and then tested whether the inclusion of climate–competition inter-
actions improved model fit. Thus, we tested models with different
ring age curves and iteratively included mast year, tree DBH, cli-
mate variables, conifer and broadleaf BA, and conifer BA in two‐
way interactions with climate variables. The model with the lowest
AICc was selected as the best “base 1”model; it included a cubic
ring age term, mast year, DBH, climate variables, broadleaf BA, and
climate–conifer BA interactions (Supporting Information Table S2).
The “base 1”model also contained two random effects: (a) mini‐
grid ID to account for spatial non‐independence of plots and (b) an
individual‐specific autoregressive (AR1) term to account for tempo-
ral autocorrelation and differences among individual trees. In the
second stage of model selection, we tested our hypothesis that
model fit would be improved by adding climate interactions with
site factors associated with drought stress as well as adding main
effects known to influence white spruce growth in DNPP (Table 1).
Thus, we tested whether including climate interactions with tree
DBH, percent tall shrub cover, plot slope angle, potential solar radi-
ation, living mat depth, or SOL improved the “base 1”model. We
tested a set of 88 competing models (Table 1, Supporting Informa-
tion Table S2). For model simplicity, we limited the number of
terms interacting with climate to two in a single model, one of
which was always conifer BA, and we excluded quadratic terms
from interactions with the aim of maintaining a sample size of
approximately 10 sampled trees per model covariate. We found the
addition of climate interactions with potential solar radiation, per-
cent shrub cover, tree DBH, living mat depth, and SOL all
improved the base model of BAI (lowered AICc); however, the
addition of potential solar radiation–climate interactions produced
the greatest reduction in AICc (Supporting Information Table S2).
Thus, we selected potential solar radiation as the second variable
(after conifer BA) to interact with climate terms. In stages 1 and 2,
we used the full suite climate variables, while in stage 3 we
assessed which of the climate variables and interactions should be
in the final model. We compared 50 models with different combi-
nations of climate variables interacted with conifer BA and
potential solar radiation (Supporting Information Table S3). We also
tested including some quadratic terms in the interaction terms. The
final selected model was the most parsimonious model with an
AICc value within two points of the lowest AICc value (Supporting
Information Table S3).
2.6
|
Relationship among climate–competition
effects, site, Δ
13
C, iWUE, and radial growth in a
subset of the DNPP tree sample
Our goals in the analyses of a subset of trees selected from a
floodplain and south‐facing hillslope were to (a) model the rela-
tionship of Δ
13
C and iWUE in tree rings to mean June–July VPD,
conifer BA, and site location (floodplain vs. hillslope), (b) determine
whether radial growth response to climate and competition in
these selected trees was similar to the overall sample, and (c)
examine the relationship between radial growth and possible
drought stress signals (Δ
13
C and iWUE). We began by modeling
Δ
13
C and iWUE as a function of mean June–July VPD interacted
with conifer BA and site. We included tree ID as a random effect.
The iWUE model additionally included a term to account for the
linear increase in iWUE over time. We then fit the same model
to log‐transformed BAI averaged over two years. We chose the
mean of BAI concurrent with and following the current growing
season year as radial growth is often related to conditions in both
the current and previous growing seasons. Finally, we modeled
the two‐year BAI average as a function of Δ
13
C interacted with
conifer BA and site and separately as a function of iWUE inter-
acted with conifer BA and site. The models of BAI included a
tree‐specific autoregressive (AR1) term to account for temporal
autocorrelation and differences among trees. Tree age can signifi-
cantly affect both growth rates as well as δ
13
C values (McCarroll
& Loader, 2004; Sullivan et al., 2017). For this reason, we tested
including ring age in each of the models (Supporting Information
Table S4). Continuous model covariates were standardized to have
a mean of zero and a standard deviation of one. Thus, model
coefficients were standardized and comparable.
We fit all mixed‐effects models using the lme4 package (Bates,
Maechler, Bolker, & Walker, 2015, version 1.1–14) in program R(R
Core Team, 2017, version 3.4.2). The correlation coefficients among
model covariates were all between −0.6 and 0.6 (Supporting Infor-
mation Figure S3). Residuals from each selected model were
assessed for violations of normality, homoscedasticity, and temporal
autocorrelation using the DHARMa package in R(Hartig, 2018).
Residuals from models conformed to all model assumptions, save
those from the full dataset BAI model, which showed some depar-
ture from normality (Supporting Information Figure S4). Variables in
the selected models were considered interpretable if the 95% confi-
dence intervals (estimate ± 1.96 ×SE of estimate) around the esti-
mate did not overlap zero or were not within 0.001 of zero. AICc
values and conditional and marginal R
2
values were calculated with
the Multi‐Model Inference package (Bartoń, 2017, version 1.40.0).
NICKLEN ET AL.
|
7
3
|
RESULTS
3.1
|
Climate–competition effects on white spruce
radial growth (full DNPP tree sample)
The addition of climate–competition interaction variables to the base
climate model significantly improved model fit as measured by AICc
(Supporting Information Table S2). The inclusion of climate interac-
tions with potential solar radiation, percent shrub cover, tree DBH,
living mat depth, and SOL all improved the base model of BAI, but
potential solar radiation–climate interactions produced the greatest
reduction in AICc (Supporting Information Table S2). Thus, in the
final model we selected potential solar radiation as the second vari-
able (after conifer BA) to interact with climate terms. The estimated
effects of covariates included in the final model on BAI are described
below.
The estimated mean white spruce BAI from the final selected
model, for the period of observation, 1974–1980 to 2003–2009,
was 330 mm
2
/year. Living mat depth, potential solar radiation, and
plot slope angle were negatively related to annual radial growth
(Table 2). Thus, an increase of +2SD above the mean for each value
would be expected to reduce annual BAI by 11%, 8%, and 7%,
respectively. Additionally, radial growth was reduced by an estimated
8% in mast years relative to non‐mast years. Radial growth was simi-
lar across lithology types except for eolian soils (loess), which were
associated with high radial growth. Unsurprisingly, tree DBH was
positively associated with radial growth. Plot conifer BA alone had
no overall influence on radial growth (but see interactive effects
below), while broadleaf BA +2SD above average was associated with
an estimated 10% increase in white spruce annual radial growth
(Table 2).
Without considering interactive effects, the climate conditions
associated with the highest estimated BAI were low mean VPD val-
ues in June–July in the year prior to ring growth, followed by wet
prior‐year August conditions and low winter precipitation, low or
high mean May VPD, and current June–July VPDs near or slightly
above average. Only winter precipitation and mean June–July VPD
FIGURE 2 Estimated white spruce
annual basal area increment (BAI, mm
2
/
year) as a function of live conifer basal
area and previous year precipitation sum in
August (a) and mean vapor pressure deficit
in June–July (VPD; b) as well as a function
of potential solar radiation and current
May precipitation sums (c), and mean May
vapor pressure deficit (d). Solar radiation,
live conifer basal area, and climate values
span the range observed within the
sample. Model covariates not shown were
held at mean values and a non‐mast year is
assumed
8
|
NICKLEN ET AL.
in year of ring formation showed no interactive effects with conifer
BA or solar radiation. Estimated BAI decreased with increasing win-
ter precipitation (Table 2). It may be that high snow cover is related
to shorter growing seasons or growing seasons shifted to later in the
spring/summer when drought stress is greater. Growth response to
current and previous June–July VPD was nonlinear. White spruce
BAI was predicted to increase with current year June–July VPD until
an estimated 7.5 hPa, above which BAI decreased (Table 2; Support-
ing Information Figure S5). In contrast, BAI was predicted to
decrease strongly with increasing previous year June–July VPD, but
at VPD values above 9 hPa BAI no longer decreased with increasing
VPD (Figure 2).
Increasing conifer BA was predicted to amplify the negative
effects of previous August drought and June–July VPDs on white
spruce radial growth (Table 2; Figure 2a). Specifically, BAI increased
from an estimated 300 mm
2
/year to over 400 mm
2
/year with
increasing precipitation levels in high conifer BA stands, while white
spruce in open stands was only predicted to increase BAI from an
estimated 325 to 350 mm
2
/year under the same August precipitation
gradient (Figure 2a). Similarly, white spruce growing in high conifer
BA stands were more sensitive to previous mean June–July VPD
than trees in low conifer BA stands (Figure 2b).
Current year May precipitation was positively associated with
BAI in high potential solar radiation (south‐facing slopes with no
sun‐obstructing topography) sites, but negatively associated with BAI
in low solar radiation sites. In years with dry May conditions, BAI
varied an estimated 75 mm
2
/year depending on site potential solar
radiation receipts, while in wet May conditions estimated radial
growth was nearly uniform across the range of site solar radiation
levels (Figure 2c). White spruce growth in low solar radiation sites
was predicted to benefit the most from high mean May VPD, while
white spruce in high solar radiation sites reduced BAI in years with
high mean May VPD (Figure 2d). White spruce in low solar radiation
sites showed the lowest estimated radial growth when May VPD
was average, while growth was higher in either cool (low VPD) or
warm (high VPD) years. Counterintuitively, previous year August pre-
cipitation benefited white spruce more in low rather than high solar
radiation sites. It is possible white spruce growth in high solar radia-
tion sites ceases for the year before these trees can benefit from
August rainfall. Alternatively, August rainfall may run off and/or
evaporates in high solar radiation sites before it is available for next
year's growth.
3.2
|
Relationship among climate–competition
effects, site, Δ
13
C, iWUE, and radial growth (subset
of DNPP tree sample)
Discrimination against
13
C(Δ
13
C) in annual tree rings was predicted
to decrease with increasing current mean June–July VPD values and
during mast years (Table 3; Figure 3), but was not influenced by esti-
mated ring age (Supporting Information Table S4). Stand competition
(conifer BA) and site (floodplain vs. hillslope) had no direct influence
on Δ
13
C levels; however, Δ
13
C was more sensitive to mean June–
July VPD in high vs. low conifer BA stands and in hillslope vs. flood-
plain trees (Table 3; Figure 3). Patterns in iWUE were the inverse of
those for Δ
13
C (Table 3).
The two‐year BAI mean (mean of current and subsequent year
BAI) was predicted to decrease with increasing current mean June–
July VPD and with mast year (Table 3). Unlike the larger DNPP sam-
ple, radial growth in this subset was directly and negatively related
TABLE 3 Standardized coefficient estimates for models
Covariates and interactions Δ
13
C model iWUE model 2‐year BAI model 2‐year BAI ~Δ13C 2‐year BAI ~iWUE
Intercept 16.487
a
105.47
a
7.261
a
7.3
a
7.33
a
Mast year −0.301
a
2.694
a
−0.084
a
−0.066
a
−0.064
a
VPD current Jun–Jul −0.455
a
4.395
a
−0.091
a
NA NA
Conifer BA 0.205 −2.038 −0.180
a
−0.14
a
−0.139
a
Site −0.066 1.085 0.685
a
0.375
a
0.382
a
Conifer BA ×VPD curr. Jun–Jul −0.160
a
1.629
a
−0.007 NA NA
Site ×VPD curr. Jun–Jul 0.301
a
−3.222
a
0.041 NA NA
Δ
13
C
b
NA NA NA 0.035 −0.059
a
Site ×Δ
13
C
b
NA NA NA −0.04 0.090
a
Conifer BA ×Δ
13
C
b
NA NA NA 0.019 −0.032
a
Marginal R
2
/conditional R
2
0.14/0.47 0.40/0.63 0.22/0.90 0.21/0.81 0.21/0.83
Notes. The first three models share the same covariates, with the first estimating Δ
13
C, the second estimating iWUE, and the third estimating log‐trans-
formed white spruce BAI (average of current and subsequent growth years). The fourth and fifth models estimate the relationship between two‐year
BAI average and Δ
13
C and iWUE, respectively. Covariates were scaled to have a mean of zero and a standard deviation of 1. NA indicates the covariate
was not in the model. Each model has tree as a random effect. BAI models additionally have a cubic ring age term that is not shown (see Supporting
Information Table S4) and an autoregressive term (AR1) by individual tree as a random effect. N= 12 trees, 360 growth rings and 247 Δ
13
C and iWUE
measurements. Estimates for “Site”are for the floodplain trees relative to the hillslope trees.
a
Estimate significant (upper and lower 95% confidence interval do not overlap zero).
b
Covariate is iWUE for 2‐year BAI~iWUE model and ×Δ
13
C for
the 2‐year BAI~Δ13C model.
NICKLEN ET AL.
|
9
to conifer BA and showed no variation in response to summer VPD
across gradients of conifer BA or site location. The average two‐year
BAI mean from the floodplain trees was an estimated 50% larger
than the hillslope trees. For context, the mean BAI from the six
floodplain and hillslope plots was 311% and 184% larger than the
mean BAI from the entire DNPP sample, respectively. These were
both relatively productive areas, with the floodplain site being the
most productive in the DNPP sample (Supporting Information Fig-
ure S1). The relationship between the two‐year BAI mean and Δ
13
C
was not significant, though there were some apparent trends. These
same trends, though in the inverse, were significant in the relation-
ship between 2‐year BAI mean and iWUE. The 2‐year BAI mean
generally decreased with increasing iWUE, but this negative trend
was amplified by stand competition and strongly dependent on site
location (Figure 3c–d). While trees in the hillslope site significantly
decreased growth with increasing iWUE, trees in the floodplain site
showed only minor growth decreases with increasing iWUE and only
at the highest stand competition levels. At low stand competition
levels, floodplain trees actually increased growth with increasing
iWUE (Figure 3c).
4
|
DISCUSSION
We found that white spruce climate‐growth response was contin-
gent on stand BA and site moisture characteristics. High competition
(stand BA) and high potential solar radiation intensified the negative
BAI response to warm and dry early to mid‐summer and dry late
summer conditions. The results of our carbon isotope analysis sup-
ported the hypothesis that moisture limitation is the mechanism for
reduced growth in warm dry years, particularly in high competition
stands and in the drier portions of the landscape. Discrimination
against
13
C diminished with high June–July VPD, and this response
was amplified in trees on south‐facing hillslopes and in high BA
stands, in keeping with our hypothesis. In productive locations
where competition for water may not be limiting, however, we
report evidence that growth is positively related to increased iWUE.
FIGURE 3 Estimated Δ
13
Casa
function of live conifer basal area and
current year mean June–July vapor
pressure deficit (VPD; a, b), and estimated
2‐year mean BAI as a function of live
conifer basal area and intrinsic water‐use
efficiency (iWUE; c, d) at the floodplain (a,
c) and south‐facing hillslope site (b, d).
Model covariates not shown were held at
mean values, and a non‐mast year is
assumed
10
|
NICKLEN ET AL.
Finally, during mast years, we found decreased radial growth,
reduced
13
C discrimination, and increased intrinsic water‐use effi-
ciency. Our findings demonstrate the significant role of temporally
variable and confounded factors, such as forest structure and cli-
mate, on the observed climate response of white spruce in interior
Alaska.
We found tree growth was interactively influenced by factors
related to succession (moss depth and conifer BA), tree aging, and
climate. This confounded and interacting nature of climate change
and other directionally changing factors on tree growth has several
implications. First, disentangling growth trends related to succes-
sional processes and tree aging from growth trends related to cli-
mate change requires a sampling design that encompasses a wide
range of tree ages and sizes growing in similar sites (Bowman, Brie-
nen, Gloor, Phillips, & Prior, 2013) and analysis procedures that
account for the influence of succession or tree aging on growth (e.g.,
utilizing appropriate detrending methods or including age‐related
growth curves and succession‐related factors in growth models). Our
findings that growth declined with increasing moss depth and
showed greater declines in high conifer BA than low conifer BA
stands in warm, dry summers suggest that if factors related to suc-
cession and tree age are not accounted for, the result will likely be
declining spruce radial growth rates over time and it will not be pos-
sible to discern whether this is due to climate changes, succession,
tree aging, or some combination thereof. This accords with the find-
ing that detrending methods influence the apparent growth trends in
white and black spruce (Sullivan et al., 2017; Sullivan, Pattison,
Brownlee, Cahoon, & Hollingsworth, 2016). It is important to note
that we did not explicitly examine trends over time in this paper, but
found that site factors that do change over time significantly influ-
ence white spruce climate‐growth responses.
Another implication of interacting climate and succession factors
is that sampling location is an important consideration when drawing
inferences from the results of tree‐ring studies. We found that white
spruce in high conifer BA stands showed a more negative growth
response and decreased Δ
13
C to June–July VPD than white spruce
in more open stands. Thus, studies conducted in mature, closed‐
canopy forests in interior Alaska may find more pronounced white
spruce growth responses to increasing summer drought than white
spruce in young, open‐canopy forests (e.g., Barber et al., 2000).
An important implication of the interactive effect of climate and
stand BA is that the maximum density of mature forests in interior
Alaska may be reduced, such that a “fully stocked”forest in a war-
mer and drier future could have a lower BA than in the past. If this
were the case, we would expect to see an increase in white spruce
mortality in high BA stands associated with warm, dry climate condi-
tions. There is mixed evidence that this reduction in stand BA is cur-
rently occurring. For example, Trugman et al. (2017) found white
spruce mortality in interior Alaska was associated with high spring
temperatures and to a lesser extent, July moisture availability, and
competition, but found no evidence of increased mortality between
1994 and 2013. It may be that this was too short a time period to
detect change in mortality rates. Indeed, longer term studies from
Canada have found increased mortality rates for white spruce and
other boreal tree species resulting from competition (Luo & Chen,
2015; Zhang et al., 2015) and drought (Peng et al., 2011). The incon-
sistent or uncertain evidence of increased white spruce mortality
may indicate that other factors are countering the negative impacts
of increasing summer drought and successional processes such as
shifting foliage to root ratios, and/or increasing atmospheric CO
2
(Angert et al., 2005; Sullivan et al., 2017). It may also be that long‐
term, competition‐induced increases in tree mortality in high basal
area stands associated with warm, dry conditions help counter the
negative impacts of a warming climate in these locations.
We show that trees in high BA sites are more sensitive to cli-
mate than those growing in more open stands. The greater climate
sensitivity of trees in mature, high competition sites may increase
vulnerability to disease and insect induced mortality (Anderegg et al.,
2015; Cahoon et al., 2018; Csank et al., 2016; Mcdowell et al.,
2008) and reduce resilience to disturbance (Johnstone, McIntire,
Pedersen, King, & Pisaric, 2010) relative to trees in younger, less
drought‐stressed stands.
Mast years, years with high cone and seed production, are
recorded as relatively narrow rings in white spruce tree‐ring series
(Juday et al., 2003). A novel and noteworthy result of our work is
that mast years are recorded as particularly low Δ
13
C and high iWUE
values in white spruce tree rings in DNPP. Similarly, current year
shoots in Fagus crenata showed
13
C enrichment during fruiting (Han,
Kagawa, Kabeya, & Inagaki, 2016). Reduced radial growth during
mast years is likely due to a shift toward reproduction at the
expense of stem growth. It is possible the reduced Δ
13
C and
increased iWUE during mast years is a result of increased photosyn-
thesis and/or greater stomatal closure resulting from the high
demand for photosynthate and water required for cone and seed
maturation. This pattern could also result from a shift in the timing
of resource allocation, with reproductive effort occurring early in
summer when water is more available and wood production occur-
ring later when temperatures are higher and drought stress is
greater. There is some evidence from temperate trees that nutrients
for cone and seed production come from multi‐year accumulated
reserves, but that carbon comes from current year photosynthesis
(Han & Kabeya, 2017). That both radial growth and Δ
13
C are
reduced in the mast year supports the idea that current year's car-
bon, and not stored carbon is being redirected from radial growth to
reproduction. Given that masting events in white spruce are climati-
cally driven (Krebs, LaMontagne, Kenney, & Boutin, 2012; Roland
et al., 2014) and exert a tax on white spruce radial growth, there
may likely be future interactions or trade‐offs between growth and
reproduction which may play out differently across the landscape
(Roland et al., 2014).
Despite the very different site conditions (floodplain vs. hillslope)
and a large gradient in conifer BA across the subset plots, the mean
Δ
13
C and iWUE were not significantly different between the flood-
plain and hillslope trees and across conifer BA levels (Table 3). This
finding is consistent with the set point theory of homeostatic gas
exchange (Brooks & Mitchell, 2011; McDowell, Adams, Bailey, Hess,
NICKLEN ET AL.
|
11
& Kolb, 2006; Whitehead, Jarvis, & Waring, 1984). Under this the-
ory, the floodplain and hillslope trees in high and low conifer BA
stands may have adjusted their architecture (e.g., leaf size, sapwood
porosity, root development) to maximize photosynthesis while mini-
mizing risks of cavitation (Fernández‐de‐Uña et al., 2016; McDowell
et al., 2006; Tyree & Sperry, 1988) to ultimately achieve a homeo-
static level of Δ
13
C and iWUE. However, during hot, dry summers
Δ
13
C decreased and iWUE increased in trees in high competition
sites relative to low competition and in hillslope relative to flood-
plain, suggesting the former trees operate on a much thinner safety
margin (lower soil water potential) and close their stomata more
quickly during warm dry periods than the latter trees. The ability of
white spruce to maintain a constant ratio between water loss and
photosynthetic gain across these disparate site conditions highlights
the phenotypic plasticity of these trees in the face of incrementally
changing conditions; however, the greater sensitivity of both growth
and Δ
13
C of the trees on high competition, south‐facing slopes
points to potential future break points in this plasticity.
In water‐limited environments, increasing summer drought is
expected to lead to stomatal closure, resulting in decreased Δ
13
C
and reduced transpiration relative to photosynthesis, and thus
increased iWUE. Prolonged stomatal closure during drought condi-
tions reduces carbon uptake (Mcdowell et al., 2008; Sala, Woodruff,
& Meinzer, 2012; Sevanto, Mcdowell, Dickman, Pangle, & Pockman,
2014). Given this, we expected to see reduced BAI with decreasing
Δ
13
C and increasing iWUE. In general, we found these expected pat-
terns, with one exception: Floodplain trees in open stands increased
growth with increasing iWUE. This finding suggests these trees may
have increased iWUE as a result of increased photosynthesis rather
than decreased stomatal conductance. This highlights the need to
interpret iWUE in concert with growth patterns and not to assume
it is metric of drought stress. We emphasize these results are from a
very small subset of trees from a productive floodplain and cannot
be extrapolated to the full landscape. Rather, these findings point to
the highly dynamic mosaic of tree growth responses to climate and
importance of stand competition and landscape position in mediating
climate‐growth responses.
We found a positive association between BAI and broadleaf BA.
This is likely driven by the site conditions associated with broadleaf
species in DNPP that are also conducive to white spruce productiv-
ity: deep active layers (Alaska birch), south‐facing slopes (aspen), and
river terraces (balsam poplar; Roland et al., 2013). Further, soil tem-
peratures and nutrient cycling may be higher in areas with high
broadleaf BA (Chapin et al., 2006). In broadleaf dominated forest
stands, high radiation input during spring thaws the ground earlier
than in conifer dominated stands (Chapin et al., 2006) and leaf litter
inhibits moss development, keeping soils warmer than in areas with
thick living mats (Roland, Stehn, Schmidt, & Houseman, 2016).
Broadleaf trees, however, take up considerably more soil water than
coniferous trees (Young‐Robertson, Bolton, Bhatt, Cristóbal, & Tho-
man, 2016) and, thus, could have a significant impact on white
spruce climate‐growth relationship. Indeed, Cortini et al. (2012)
found this to be the case in mixedwood forests in western Canada.
Because broadleaved trees constitute a small fraction of the DNPP
forest mosaic, we were unable to test the effect of broadleaf BA on
white spruce growth response to climate. This should be a focus of
future work in study areas with greater broadleaf presence and rep-
resentation in different site conditions than found DNPP.
It is not surprising that we found that large diameter trees
showed higher annual BAI than small diameter trees. It is worth not-
ing, however, that tree DBH had an effect size two to forty times
larger than the climate and interaction variables. Similarly, ring age
also had a nearly twofold to 25‐fold larger effect than climate and
interaction variables. Our model results are consistent with the find-
ing that mass growth rate increases with individual tree size (Foster,
Finley, D'Amato, Bradford, & Banerjee, 2016; Stephenson et al.,
2014) and that tree size and age have larger impacts on tree produc-
tivity within a region than climate (Foster et al., 2016). The strong
positive influence of tree size on growth underscores the need to
interpret climate‐growth responses from within the context of cur-
rent forest physical structure. If predictions of forest productivity or
carbon storage were based only on dendroecological studies using
dimensionless ring width indices to determine climate‐growth, the
large and significant effect of current tree size and age would be lost
resulting in exaggerated climate effects on future productivity. There
is likely some interactive effect between climate and tree size as has
been found with climate and tree age (Carrer & Urbinati, 2011; Sze-
icz & MacDonald, 1994). We found including climate–DBH interac-
tive effects in our growth model improved the model fit (Supporting
Information Table S2), but for simplicity, we only included interac-
tions with the one variable in addition to conifer basal area that best
improved the model fit; thus, climate–DBH interactions were not
included in our final growth model. Our understanding of white
spruce growth in interior Alaska would benefit from an explicit
examination of climate and tree size interactions.
There are several limitations to our models of white spruce
growth. First, our estimates of growth are limited to stem growth.
Thus, we are unable to determine whether a change in radial growth
represents an overall change in productivity or a reallocation of car-
bon among stems, roots, branch elongation, or needles. We did,
however, include reproductive effort into our growth, Δ
13
C, and
iWUE models, which revealed significant trade‐offs between radial
growth and reproduction as well as distinct Δ
13
C and iWUE patterns
associated with masting. Second, although our study design ensured
a random sample of plot locations across the study area, at the plot
level our cored trees tended to be slightly larger than the average
tree within the plot (Supporting Information Figure S2) indicating a
tree size bias. This bias may have led to an underestimated competi-
tion effect in our model, assuming larger trees are less negatively
affected by competition with smaller trees and perhaps to an overes-
timated climate effect, assuming larger trees are more drought sensi-
tive than smaller trees (Bennett et al., 2015). Finally, we did not core
dead trees, so our sample may be affected by “modern sample bias.”
However, because we limited the time frame of our study to
30 years and our sample includes a large range in tree ages from a
random placement of plots, with relatively few dead trees, our
12
|
NICKLEN ET AL.
sample should be relatively robust and buffered from some of the
biases associated with sampling only living trees.
In summary, we found stand BA mediates the influence of cli-
mate on the annual radial growth of white spruce in DNPP, amplify-
ing the negative effect of previous summer VPD and moderating the
positive influence of previous August rainfall. Our carbon isotope
analysis suggests the mechanism behind these modified climate‐
growth responses may be increased competition for moisture in high
basal area stands and dry sites. We also found that large reproduc-
tive events (mast years) both reduce radial growth and strongly
decrease Δ
13
C (and increase iWUE) of white spruce trees, suggesting
trade‐offs between growth and reproduction for current year's pho-
tosynthate. Our finding that high BA stands show greater sensitivity
and negative growth responses to warming climate conditions than
open stands may ultimately portend lower white spruce stand densi-
ties and increased vulnerability to insects and disease in future inte-
rior Alaska mature forests. Our findings also point to the need for
studies examining growth trends to address the confounded nature
of climate change and other directionally changing factors that influ-
ence tree growth (succession, tree aging, atmospheric CO
2
). Overall,
our results suggest highly dynamic individual tree growth responses
to future climate change that are dependent on both landscape posi-
tion and stand competition and likely to result in feedbacks on
future forest structure.
ACKNOWLEDGEMENTS
We thank the many people who collected field data for the Central
Alaska Network vegetation monitoring program as well as those who
helped with database design and data management. We are grateful
to Dr. A. Lloyd's laboratory for measuring and cross‐dating many of
the cores used in this analysis. We thank J. Schmidt and four anony-
mous reviewers for insightful comments that greatly improved this
manuscript. The Central Alaska Network and Denali National Park
and Preserve as part of the U.S. National Park Service Inventory and
Monitoring Program funded this project.
ORCID
Elizabeth Fleur Nicklen https://orcid.org/0000-0001-5489-807X
REFERENCES
Aakala, T., Fraver, S., D’Amato, A. W., & Palik, B. J. (2013). Influence of
competition and age on tree growth in structurally complex old‐growth
forests in northern Minnesota, USA. Forest Ecology and Management,
308, 128–135. https://doi.org/10.1016/j.foreco.2013.07.057
Alam, S. A., Huang, J.‐G., Stadt, K. J., Comeau, P. G., Dawson, A., Gea‐
Izquierdo, G., …Berninger, F. (2017). Effects of competition, drought
stress and photosynthetic productivity on the radial growth of white
spruce in Western Canada. Frontiers in Plant Science,8,1–15.
https://doi.org/10.3389/fpls.2017.01915
Anchukaitis, K. J., Evans, M. N., Lange, T., Smith, D. R., Leavitt, S. W., &
Schrag, D. P. (2008). Consequences of a rapid cellulose extraction
technique for oxygen isotope and radiocarbon analyses. Analytical
Chemistry,80, 2035–2041. https://doi.org/10.1021/ac7020272
Anderegg, W. R., Hicke, J. A., Fisher, R. A., Allen, C. D., Aukema, J., Bentz,
B., …Zeppel, M. (2015). Tree mortality from drought, insects, and
their interactions in a changing climate. New Phytologist,208, 674–
683. https://doi.org/10.1111/nph.13477
Angert, A., Biraud, S., Bonfils, C., Henning, C. C., Buermann, W., Pinzon,
J., …Fung, I. (2005). Drier summers cancel out the CO
2
uptake
enhancement induced by warmer springs. Proceedings of the National
Academy of Sciences,102, 10823–10827. https://doi.org/10.1073/
pnas.0501647102
Barber, V. A., Juday, G. P., & Finney, B. P. (2000). Reduced growth of
Alaskan white spruce in the 20th century from temperature‐induced
drought stress. Nature,405, 668–673.
Barrett, K., McGuire, A. D., Hoy, E. E., & Kasischke, E. S. (2011). Potential
shifts in dominant forest cover in interior Alaska driven by variations
in fire severity. Ecological Applications,21, 2380–2396. https://doi.
org/10.1890/10-0896.1
Bates, D., Maechler, M., Bolker, B., & Walker, S. (2015). Fitting linear
mixed‐effects models using lme4. Journal of Statistical Software,67,
1–48.
Beck, P. S., Juday, G. P., Alix, C., Barber, V. A., Winslow, S. E., Sousa, E.
E., …Goetz, S. J. (2011). Changes in forest productivity across Alaska
consistent with biome shift. Ecology Letters,14, 373–379. https://doi.
org/10.1111/j.1461-0248.2011.01598.x
Bennett, A. C., Mcdowell, N. G., Allen, C. D., & Anderson‐Teixeira, K. J.
(2015). Larger trees suffer most during drought in forests worldwide.
Nature Plants,1,1–5. https://doi.org/10.1038/nplants.2015.139
Betts, A. K., & Ball, J. H. (1997). Albedo over the boreal forest. Journal of
Geophysical Research,102, 901–909.
Bieniek, P. A., Bhatt, U. S., Thoman, R. L., Angeloff, H., Partain, J., Pap-
ineau, J., …Gens, R. (2012). Climate divisions for Alaska based on
objective methods. Journal of Applied Meteorology and Climatology,
51, 1276–1289. https://doi.org/10.1175/JAMC-D-11-0168.1
Bonan, G. B. (2008). Forests and climate change: Forcings, feedbacks,
and the climate benefits of forests. Science,320, 1444–1449.
https://doi.org/10.1126/science.1155121
Bowman, D. M. J. S., Brienen, R. J. W., Gloor, E., Phillips, O. L., & Prior,
L. D. (2013). Detecting trends in tree growth: Not so simple. Trends
in Plant Science,18,11–17. https://doi.org/10.1016/j.tplants.2012.08.
005
Brendel, O., Iannetta, P. P. M., & Stewart, D. (2000). A rapid and simple
method to isolate pure alpha‐cellulose. Phytochemical Analysis,11,7–
10. https://doi.org/10.1002/(SICI)1099-1565(200001/02)11:1<7:
AID-PCA488>3.0.CO;2-U
Brooks, J. R., & Mitchell, A. K. (2011). Interpreting tree responses to thin-
ning and fertilization using tree‐ring stable isotopes. New Phytologist,
190, 770–782. https://doi.org/10.1111/j.1469-8137.2010.03627.x
Bunn, A. G. (2008). A dendrochronology program library in R (dplR). Den-
drochronologia,26, 115–124. https://doi.org/10.1016/j.dendro.2008.
01.002
Bunn, A. G., & Goetz, S. J. (2006). Trends in satellite‐observed circumpo-
lar photosynthetic activity from 1982 to 2003: The influence of sea-
sonality, cover type, and vegetation density. Earth Interactions,10,1–
19.
Bunn, A., & Korpela, M. (2016). Crossdating in dplR, vol. 2,1–12.
Cahoon, S. M. P., Sullivan, P. F., Brownlee, A. H., Pattison, R. R., Ander-
sen, H.‐E., Legner, K., & Hollingsworth, T. N. (2018). Contrasting dri-
vers and trends of coniferous and deciduous tree growth in interior
Alaska. Ecology,99, 1284–1295. https://doi.org/10.1002/ecy.2223
Carrer, M., & Urbinati, C. (2011). Age‐dependent tree‐ring growth
responses to climate in Larix decidua and Pinus cembra.Ecology,85,
730–740. https://doi.org/10.1890/02-0478
Chapin, F. S., McFarland, J., David McGuire, A., Euskirchen, E. S., Ruess,
R. W., & Kielland, K. (2009). The changing global carbon cycle:
NICKLEN ET AL.
|
13
Linking plant‐soil carbon dynamics to global consequences. Journal of
Ecology,97, 840–850. https://doi.org/10.1111/j.1365-2745.2009.
01529.x
Chapin, F. S. I., Viereck, L., Adams, P., Van Cleve, K., Fastie, C., Ott, R. A.,
…Johnstone, J. F. (2006). Successional processes in the Alaskan bor-
eal forest. In F. I. Chapin, M. W. Oswood, K. Van Cleve, L. A. Viereck,
& D. L. Verbyla (Eds.), Alaska’s changing boreal forest (pp. 100–120).
Oxford, UK: Oxford University Press.
Clark, M. H., & Duffy, M. S. (2006). Soil survey of Denali National Park
area Alaska. Alaska: Palmer.
Coomes, D. A., & Allen, R. B. (2007). Effects of size, competition and alti-
tude on tree growth. Journal of Ecology,95, 1084–1097. https://doi.
org/10.1111/j.1365-2745.2007.01280.x
Cortini, F., & Comeau, P. G. (2008). Evaluation of competitive effects of
green alder, willow and other tall shrubs on white spruce and lodge-
pole pine in Northern Alberta. Forest Ecology and Management,255,
82–91. https://doi.org/10.1016/j.foreco.2007.08.027
Cortini, F., Comeau, P. G., & Bokalo, M. (2012). Trembling aspen competi-
tion and climate effects on white spruce growth in boreal mixtures of
Western Canada. Forest Ecology and Management,277,67–73.
https://doi.org/10.1016/j.foreco.2012.04.022
Cox, P. M., Betts, R. A., Jones, C. D., Spall, S. A., & Totterdell, I. J. (2000).
Acceleration of global warming due to carbon‐cycle feedbacks in a
coupled climate model. Nature,408, 184–187.
Csank, A. Z., Miller, A. E., Sherriff, R. L., Berg, E. E., & Welker, J. M.
(2016). Tree‐ring isotopes reveal drought sensitivity in trees killed by
spruce beetle outbreaks in south‐central Alaska. Ecological Applica-
tions,26, 2001–2020. https://doi.org/10.1002/eap.1365
D’Arrigo, R. D., Kaufmann, R. K., Davi, N., Jacoby, G. C., Laskowski, C.,
Myneni, R. B., & Cherubini, P. (2004). Thresholds for warming‐in-
duced growth decline at elevational tree line in the Yukon Territory,
Canada. Global Biogeochemical Cycles,18,1–7. https://doi.org/10.
1029/2004GB002249
Daly, C. (2009). Alaska PRISM Climate Maps 1971–2000.
Daly, C., Halbleib, M., Smith, J. I., Gibson, W. P., Doggett, M. K., Taylor,
G. H., …Pasteris, P. P. (2008). Physiographically sensitive mapping of
climatological temperature and precipitation across the conterminous
United States. International Journal of Climatology,28, 2031–2064.
https://doi.org/10.1002/joc.1688
Drobyshev, I., Simard, M., Bergeron, Y., & Hofgaard, A. (2010). Does soil
organic layer thickness affect climate‐growth relationships in the
black spruce boreal ecosystem? Ecosystems,13, 556–574. https://doi.
org/10.1007/s10021-010-9340-7
Dubayah, R., & Rich, P. M. (1995). Topographic solar radiation models for
GIS. Journal of Geographic Information Systems,9, 405–413. https://d
oi.org/10.1080/02693799508902046
Euskirchen, E. S., McGuire, A. D., Rupp, T. S., Chapin, F. S., & Walsh, J. E.
(2009). Projected changes in atmospheric heating due to changes in
fire disturbance and the snow season in the western Arctic, 2003–
2100. Journal of Geophysical Research: Biogeosciences,114,1–15.
https://doi.org/10.1029/2009JG001095
Farquhar, G. D., Hubick, K. T., Condon, A. G., & Richards, R. A. (1989).
Carbon isotope fractionation and plant water‐use efficiency. In P. W.
Rundel, J. R. Ehleringer, & K. A. Nagy (Eds.), Stable isotopes in ecologi-
cal research. New York, NY: Springer.
Farquhar, G., O’Leary, M., & Berry, J. (1982). On the relationship
between carbon isotope discrimination and the intercellular carbon
dioxide concentration in leaves. Australian Journal of Plant Physiology,
9, 121–137. https://doi.org/10.1071/PP9820121
Fernández‐de‐Uña, L., Cañellas, I., & Gea‐Izquierdo, G. (2015). Stand com-
petition determines how different tree species will cope with a
warming climate. PLoS ONE,10,1–18. https://doi.org/10.1371/jour
nal.pone.0122255
Fernández‐de‐Uña, L., McDowell, N. G., Cañellas, I., & Gea‐Izquierdo, G.
(2016). Disentangling the effect of competition, CO
2
and climate on
intrinsic water‐use efficiency and tree growth. Journal of Ecology,
104, 678–690.
Foster, J. R., Finley, A. O., D’Amato, A. W., Bradford, J. B., & Banerjee, S.
(2016). Predicting tree biomass growth in the temperate‐boreal eco-
tone: Is tree size, age, competition, or climate response most impor-
tant? Global Change Biology,22, 2138–2151. https://doi.org/10.
1111/gcb.13208
Francey, R. J., & Farquhar, G. D. (1982). An explanation of 13C/12C vari-
ations in tree rings. Nature,297,28–31. https://doi.org/10.1038/
297028a0
Gewehr, S., Drobyshev, I., Berninger, F., & Bergeron, Y. (2014). Soil char-
acteristics mediate the distribution and response of boreal trees to
climatic variability. Canadian Journal of Forest Research,44, 487–498.
https://doi.org/10.1139/cjfr-2013-0481
Girardin, M. P., Bouriaud, O., Hogg, E. H., Kurz, W., Zimmermann, N. E.,
Metsaranta, J. M., …Bhatti, J. (2016). No growth stimulation of
Canada’s boreal forest under half‐century of combined warming and
CO
2
fertilization. Proceedings of the National Academy of Sciences,
113, E8406–E8414.
Han, Q., & Kabeya, D. (2017). Recent developments in understanding
mast seeding in relation to dynamics of carbon and nitrogen
resources in temperate trees. Ecological Research,32, 771–778.
https://doi.org/10.1007/s11284-017-1494-8
Han, Q., Kagawa, A., Kabeya, D., & Inagaki, Y. (2016). Reproduction‐re-
lated variation in carbon allocation to woody tissues in Fagus crenata
using a natural 13 C approach. Tree Physiology,36, 1343–1352.
Hartig, F. (2018). DHARMa: Residual Diagnostics for Hierarchical (Multi‐
Level / Mixed) Regression Models. R package version 0.2.0.
Hellmann, L., Agafonov, L., Charpentier Ljungqvist, F., Churakova, O.,
Düthorn, E., Esper, J., …Büntgen, U. (2016). Diverse growth trends
and climate responses across Eurasia’s boreal. Environmental Research
Letters,11, 074021.
Hinzman, L. D., Bettez, N. D., Bolton, W. R., Chapin, F. S., Dyurgerov, M.
B., Fastie, C. L., …Yoshikawa, K. (2005). Evidence and implications
of recent climate change in Northern Alaska and other Arctic regions.
Climatic Change,72, 251–298. https://doi.org/10.1007/s10584-005-
5352-2
Holzkämper, S., Tillman, P. K., Kuhry, P., & Esper, J. (2012). Comparison
of stable carbon and oxygen isotopes in Picea glauca tree rings and
Sphagnum fuscum moss remains from subarctic Canada. Quaternary
Research (United States),78, 295–302. https://doi.org/10.1016/j.yqre
s.2012.05.014
Huang, J., Tardif, J. C., Bergeron, Y., Denneler, B., Berninger, F., & Girar-
din, M. P. (2010). Radial growth response of four dominant boreal
tree species to climate along a latitudinal gradient in the eastern
Canadian boreal forest. Global Change Biology,16, 711–731. https://d
oi.org/10.1111/j.1365-2486.2009.01990.x
IPCC. (2014). Climate Change 2014: Synthesis Report. In Core Writing
Team, R. K. Pachauri, & L. A. Meyer (Eds.), Contribution of working
groups I, II and III to the fifth assessment report of the intergovern-
mental panel on climate change (pp. 1–112). Geneva, Switzerland:
IPCC.
Jiang, Y., Zhang, J., Han, S., Chen, Z., Setälä, H., Yu, J., …Gu, Y. (2016).
Radial growth response of Larix gmelinii to climate along a latitudinal
gradient in the greater Khingan mountains, northeastern China. For-
ests,7,1–12. https://doi.org/10.3390/f7120295
Johnstone, J. F., Hollingsworth, T. N., Chapin, F. S., & Mack, M. C.
(2010). Changes in fire regime break the legacy lock on successional
trajectories in Alaskan boreal forest. Global Change Biology,16, 1281–
1295.
Johnstone, J. F., McIntire, E. J. B., Pedersen, E. J., King, G., & Pisaric, M.
J. F. (2010). A sensitive slope: Estimating landscape patterns of forest
resilience in a changing climate. Ecosphere,1,1–21.
Juday, G. P., & Alix, C. (2012). Consistent negative temperature sensitiv-
ity and positive influence of precipitation on growth of floodplain
14
|
NICKLEN ET AL.
Picea glauca in Interior Alaska. Canadian Journal of Forest Research,
42, 561–573.
Juday, G. P., Alix, C., & Grant, T. A. (2015). Spatial coherence and change of
opposite white spruce temperature sensitivities on floodplains in
Alaska confirms early‐stage boreal biome shift. Forest Ecology and Man-
agement,350,46–61. https://doi.org/10.1016/j.foreco.2015.04.016
Juday, G. P., Barber, V. A., Rupp, S. T., Zasada, J. C., & Wilmking, M.
(2003). A 200 year perspective of climate variability and the response
of white spruce in interior Alaska. In Climate variability and ecosystem
response at longterm ecological research sites (pp. 226–250). Oxford:
Oxford University Press.
Kasischke, E. S., Williams, D., & Barry, D. (2002). Analysis of the patterns
of large fires in the boreal forest region of Alaska. International Jour-
nal of Wildland Fire,11, 131–144.
Keeling, C. D., Piper, S. C., Bacastow, R. B., Wahlen, M., Whorf, T. P.,
Heimann, M., & Meijer, H. A. (2001). Exchanges of atmospheric CO
2
and 13CO
2
with the terrestrial biosphere and oceans from 1978 to
2000. San Diego, 88 pp.
Koven, C. D. (2013). Boreal carbon loss due to poleward shift in low‐car-
bon ecosystems. Nature Geoscience,6, 452–456. https://doi.org/10.
1038/ngeo1801
Krebs, C. J., LaMontagne, J. M., Kenney, A. J., & Boutin, S. (2012). Cli-
matic determinants of white spruce cone crops in the boreal forest
of southwestern Yukon. Botany‐Botanique,90, 113–119. https://doi.
org/10.1139/b11-088
Kuusela, K. (1990). The dynamics of boreal coniferous forests. Helsinki, Fin-
land: Finnish National Fund for Research and Development (SITRA).
Lloyd, A. H., & Bunn, A. G. (2007). Responses of the circumpolar boreal
forest to 20th century climate variability. Environmental Research Let-
ters,2, 045013. https://doi.org/10.1088/1748-9326/2/4/045013
Lloyd, A. H., Duffy, P. A., & Mann, D. H. (2013). Nonlinear responses of
white spruce growth to climate variability in interior Alaska. Canadian
Journal of Forest Research,43, 331–343. https://doi.org/10.1139/cjfr-
2012-0372
Luo, Y., & Chen, H. Y. H. (2015). Climate change‐associated tree mortality
increases without decreasing water availability. Ecology Letters,18,
1207–1215. https://doi.org/10.1111/ele.12500
Man, R., & Lieffers, V. J. (1999). Are mixtures of aspen and white spruce
more productive than single species stands? Forestry Chronicle,75,
505–513. https://doi.org/10.5558/tfc75505-3
Martin, G. L., & Ek, A. R. (1984). A comparison of competition measures
and growth models for prediction plantation Red Pine diameter and
height growth. Forest Science,30, 731–743.
McAfee, S., Guentchev, G., & Eischeid, J. (2014). Reconciling precipitation
trends in Alaska: 2. Gridded data analyses. Journal of Geophysical
Research: Atmospheres,119, 13820–13837.
McCarroll, D., & Loader, N. J. (2004). Stable isotopes in tree rings. Qua-
ternary Science Reviews,23, 771–801. https://doi.org/10.1016/j.quasc
irev.2003.06.017
McDowell, N. G., Adams, H. D., Bailey, J. D., Hess, M., & Kolb, T. E.
(2006). Homeostatic maintenance of ponderosa pine gas exchange.
Ecological Applications,16, 1164–1182.
Mcdowell, N. G., Pockman, W. T., Allen, C. D., Breshears, D. D., Cobb,
N., Kolb, T., …Yepez, E. A. (2008). Mechanisms of plant survival and
mortality during drought: Why do some plants survive while others
succumb to drought? New Phytologist,178, 719–739. https://doi.org/
10.1111/j.1469-8137.2008.02436.x
McGuire, A. D., Ruess, R. W., Lloyd, A., Yarie, J., Clein, J. S., & Juday, G.
P. (2010). Vulnerability of white spruce tree growth in interior Alaska
in response to climate variability: Dendrochronological, demographic,
and experimental perspectives. Canadian Journal of Forest Research,
40, 1197–1209.
Murray, F. W. (1967). On the computation of saturation vapor pressure.
Journal of Applied Meteorology and Climatology,6, 203–204. https://
doi.org/10.1175/1520-0450(1967)006<0203:OTCOSV>2.0.CO;2
Nicklen, E. F., Roland, C. A., Ruess, R. W., Schmidt, J. H., & Lloyd, A.
H. (2016). Local site conditions drive climate‐growth responses of
Picea mariana and Picea glauca in interior Alaska. Ecosphere,7,1–
34.
Ohse, B., Jansen, F., & Wilmking, M. (2012). Do limiting factors at Alas-
kan treelines shift with climatic regimes? Environmental Research Let-
ters,7, 015505.
Peng, C., Ma, Z., Lei, X., Zhu, Q., Chen, H., Wang, W., …Zhou, X. (2011).
A drought‐induced pervasive increase in tree mortality across Cana-
da’s boreal forests. Nature Climate Change,1, 467–471. https://doi.
org/10.1038/nclimate1293
Piutti, E., & Cescatti, A. (1997). A quantitative analysis of the interactions
between climatic response and intraspecific competition in European
beech. Canadian Journal of Forest Research,27, 277–284. https://doi.
org/10.1139/x96-176
Porter, T. J., Pisaric, M. F. J., Kokelj, S. V., & Edwards, T. W. D. (2009).
Climatic Signals in δ
13
C and δ
18
O of Tree‐rings from White Spruce in
the Mackenzie Delta Region, Northern Canada. Arctic, Antarctic, and
Alpine Research,41, 497–505.
R Core Team. (2017). R: A language and environment for statistical com-
puting. Vienna, Austria: R Foundation for Statistical Computing.
https://www.R‐project.org/.
Rich, P. M., Dubayah, R., Hetrick, W. A., & Saving, S. C. (1994). Using
viewshed models to calculate intercepted solar radiation: applications
in ecology. Photogrammetry and Remote Sensing Technical Papers.
American Society for Photogrammetry and Remote Sensing,
Bethesda, Maryland, USA, pp. 524–552.
Roland, C. A., Oakley, K., Debevec, E. M., & Loomis, T. (2004). Monitoring
vegetation structure and composition at multiple scales in the Central
Alaska Network. Alaska, USA: Fairbanks.
Roland, C. A., Schmidt, J. H., & Johnstone, J. F. (2014). Climate sensitivity
of reproduction in a mast‐seeding boreal conifer across its distribu-
tional range from lowland to treeline forests. Oecologia,174, 665–
677. https://doi.org/10.1007/s00442-013-2821-6
Roland, C. A., Schmidt, J. H., & Nicklen, E. F. (2013). Landscape‐scale pat-
terns in tree occupancy and abundance in subarctic Alaska. Ecological
Monographs,83,19–48. https://doi.org/10.1890/11-2136.1
Roland, C. A., Stehn, S. E., Schmidt, J., & Houseman, B. (2016). Proliferat-
ing poplars: The leading edge of landscape change in an Alaskan sub-
alpine chronosequence. Ecosphere,7,1–30. https://doi.org/10.1002/
ecs2.1398
Ruiz‐Benito, P., Madrigal‐González, J., Ratcliffe, S., Coomes, D. A., Känd-
ler, G., Lehtonen, A., …Zavala, M. A. (2014). Stand structure and
recent climate change constrain stand basal area change in European
forests: A comparison across boreal, temperate, and mediterranean
biomes. Ecosystems,17, 1439–1454. https://doi.org/10.1007/
s10021-014-9806-0
Sala, A., Woodruff, D. R., & Meinzer, F. C. (2012). Carbon dynamics in
trees: Feast or famine? Tree Physiology,32, 764–775. https://doi.org/
10.1093/treephys/tpr143
Sevanto, S., Mcdowell, N. G., Dickman, L. T., Pangle, R., & Pockman, W.
T. (2014). How do trees die? A test of the hydraulic failure and car-
bon starvation hypotheses. Plant, Cell and Environment,37, 153–161.
https://doi.org/10.1111/pce.12141
Shaver, G., Billings, W., Chapin, F. III, Giblin, A., & Nadelhoffer, K. (1992).
Global change and the carbon balance of arctic ecosystems changes
in global terrestrial carbon cycling. BioScience,42, 433–441. https://d
oi.org/10.2307/1311862
Sousanes, P. J. (2008). Annual climate summary 2006. Central Alaska
Network. Natural Resource Technical Report NPS/CAKN/NRTR‐
2008/141. Fort Collins, Colorado, USA.
Stephenson, N. L., Das, A. J., Condit, R., Russo, S. E., Baker, P. J., Beck-
man, N. G., …Zavala, M. A. (2014). Rate of tree carbon accumulation
increases continuously with tree size. Nature,507,90–93. https://d
oi.org/10.1038/nature12914
NICKLEN ET AL.
|
15
Sullivan, P. F., Pattison, R. R., Brownlee, A. H., Cahoon, S. M. P., & Hol-
lingsworth, T. N. (2016). Effect of tree‐ring detrending method on
apparent growth trends of black and white spruce in interior Alaska.
Environmental Research Letters,11,1–12. https://doi.org/10.1088/
1748-9326/11/11/114007
Sullivan, P. F., Pattison, R. R., Brownlee, A. H., Cahoon, S. M. P., & Hol-
lingsworth, T. N. (2017). Limited evidence of declining growth among
moisture‐limited black and white spruce in interior Alaska. Scientific
Reports,7,1–14. https://doi.org/10.1038/s41598-017-15644-7
Sullivan, P. F., & Sveinbjörnsson, B. (2011). Environmental controls on
needle gas exchange and growth of white spruce (Picea glauca)ona
riverside terrace near the arctic treeline. Arctic, Antarctic, and Alpine
Research,43, 279–288.
Szeicz, J. M., & MacDonald, G. M. (1994). Age‐dependent tree‐ring
growth responses of subarctic white spruce to climate. Canadian
Journal of Forest Research,24, 120–132. https://doi.org/10.1139/
x94-017
Trugman, A. T., Medvigy, D., Anderegg, W. R. L., & Pacala, S. W. (2017).
Differential declines in Alaskan boreal forest vitality related to climate
and competition. Global Change Biology,24, 1097–1107. https://doi.
org/10.1111/gcb.13952
Tyree, M. T., & Sperry, J. S. (1988). Do woody plants operate near the
point of catastrophic xylem dysfunction caused by dynamic water
stress?: Answers from a model. Plant Physiology,88, 574–580.
van Mantgem, P. J., Stephenson, N. L., Byrne, J. C., Daniels, L. D., Frank-
lin, J. F., Fulé, P. Z., …Veblen, T. T. (2009). Widespread increase of
tree mortality rates in the Western United States. Science,323, 521–
524. https://doi.org/10.1126/science.1165000
Walker, X., & Johnstone, J. F. (2014). Widespread negative correlations
between black spruce growth and temperature across topographic
moisture gradients in the boreal forest. Environmental Research Let-
ters,9, 064016. https://doi.org/10.1088/1748-9326/9/6/064016
Whitehead, D., Jarvis, P. G., & Waring, R. H. (1984). Stomatal conduc-
tance, transpiration, and resistance to water uptake in a Pinus-
sylvestris spacing experiment. Canadian Journal of Botany,14, 692–
700.
Wilmking, M., Juday, G. P., Barber, V. A., & Zald, H. S. J. (2004). Recent
climate warming forces contrasting growth responses of white spruce
at treeline in Alaska through temperature thresholds. Global Change
Biology,10, 1724–1736. https://doi.org/10.1111/j.1365-2486.2004.
00826.x
Wilmking, M., Juday, G. P., Terwilliger, M., & Barber, V. A. (2006). Model-
ing spatial variability of white spruce (Picea glauca) growth responses
to climate change at and below treeline in Alaska ‐a case study from
two national parks. Erdkunde,2, 113–126. https://doi.org/10.3112/
erdkunde.2006.02.03
Wright, M., Sherriff, R. L., Miller, A. E., & Wilson, T. (2018). Stand basal
area and temperature interact to influence growth in white spruce in
southwest Alaska. Ecosphere,9, e02462. https://doi.org/10.1002/
ecs2.2462
Yarie, J., & Van Cleve, K. (2010). Long‐term monitoring of climatic and
nutritional affects on tree growth in interior Alaska. Canadian Journal
of Forest Research,40, 1325–1335. https://doi.org/10.1139/x10-114
Yarie, J., Van Cleve, K., & Schlentner, R. (1990). Interaction between
moisture, nutrients and growth of white spruce in interior Alaska.
Forest Ecology and Management,30,73–89. https://doi.org/10.1016/
0378-1127(90)90128-X
Young‐Robertson, J. M., Bolton, W. R., Bhatt, U. S., Cristóbal, J., & Tho-
man, R. (2016). Deciduous trees are a large and overlooked sink for
snowmelt water in the boreal forest. Scientific Reports,6,1–10.
https://doi.org/10.1038/srep29504
Zhang, J., Huang, S., & He, F. (2015). Half‐century evidence from western
Canada shows forest dynamics are primarily driven by competition
followed by climate. Proceedings of the National Academy of Sciences,
112, 4009–4014. https://doi.org/10.1073/pnas.1420844112
SUPPORTING INFORMATION
Additional supporting information may be found online in the
Supporting Information section at the end of the article.
How to cite this article: Nicklen EF, Roland CA, Csank AZ,
Wilmking M, Ruess RW, Muldoon LA. Stand basal area and
solar radiation amplify white spruce climate sensitivity in
interior Alaska: Evidence from carbon isotopes and tree rings.
Glob Change Biol. 2018;00:1–16. https://doi.org/10.1111/
gcb.14511
16
|
NICKLEN ET AL.
