Content uploaded by Monica T. Rother
Author content
All content in this area was uploaded by Monica T. Rother on Sep 17, 2018
Content may be subject to copyright.
ARTICLE
A field experiment informs expected patterns of conifer
regeneration after disturbance under changing climate
conditions
Monica T. Rother, Thomas T. Veblen, and Luke G. Furman
Abstract: Climate change may inhibit tree regeneration following disturbances such as wildfire, altering post-disturbance vegetation
trajectories. We implemented a field experiment to examine the effects of manipulations of temperature and water on ponderosa
pine (Pinus ponderosa Douglas ex P. Lawson & C. Lawson) and Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) seedlings planted in a
low-elevation, recently disturbed setting of the Colorado Front Range. We implemented four treatments: warmed only (Wm), watered
only (Wt), warmed and watered (WmWt), and control (Co). We found that measures of growth and survival varied significantly by
treatment type. Average growth and survival was highest in the Wt plots, followed by the Co, WmWt, and Wm plots, respectively. This
general trend was observed for both conifer species, although average growth and survival was generally higher in ponderosa pine
than in Douglas-fir. Our findings suggest that warming temperatures and associated drought are likely to inhibit post-disturbance
regeneration of ponderosa pine and Douglas-fir in low-elevation forests of the Colorado Front Range and that future vegetation
composition and structure may differ notably from historic patterns in some areas. Our findings are relevant to other forested
ecosystems in which a warming climate may similarly inhibit regeneration by dominant tree species.
Key words: field experiment, tree regeneration, climate change, wildfire, ponderosa pine, open-top chambers.
Résumé : Les changements climatiques pourraient inhiber la régénération des arbres a
`la suite de perturbations telles qu'un feu,
ce qui modifierait les trajectoires de la végétation après une perturbation. Nous avons effectué une expérience de terrain pour
étudier les effets de manipulations de la température et de l'eau sur la croissance et la survie des semis de pin ponderosa (Pinus
ponderosa Douglas ex P. Lawson & C. Lawson) et de douglas de Menzies (Pseudotsuga menziesii (Mirb.) Franco) plantés a
`faible
altitude dans un milieu récemment perturbé du Colorado Front Range (CFR). Nous avons établi quatre traitements : réchauffe-
ment seulement (Wm), arrosage seulement (Wt), réchauffement et arrosage (WmWt) et témoin (Co). Nous avons observé que les
mesures de croissance et de survie variaient significativement selon le traitement. La croissance et la survie étaient en moyenne
les plus élevées dans les parcelles Wt suivies respectivement des parcelles Co, WmWt, et Wm. Cette tendance générale a été
observée chez les deux espèces de conifère même si la croissance et la survie étaient en général meilleures chez le pin ponderosa
que chez le douglas de Menzies. Nos résultats indiquent que l'augmentation de la température et la sécheresse qui y est associée
vont probablement inhiber la régénération qui suit une perturbation chez le pin ponderosa et le douglas de Menzies dans
les forêts du CFR situées a
`faible altitude et que la composition et la structure de la régénération pourraient dans le futur être très
différentes de la configuration historique dans certaines régions. Bien que notre étude mette l'accent sur les forêts du CFR situées
a
`faible altitude nos résultats sont pertinents pour d'autres écosystèmes forestiers où le réchauffement du climat peut de façon
similaire inhiber la régénération des espèces dominantes. [Traduit par la Rédaction]
Mots-clés : expérience de terrain, régénération des arbres, changement climatique, feu de forêt, pin ponderosa, chambres a
`ciel
ouvert.
Introduction
Recent studies of vegetation patterns following wildfire have
been motivated by concern that climate change and (or) potential
increases in wildfire severity may alter postfire vegetation trajec-
tories by inhibiting processes of conifer regeneration. In some dry
ponderosa pine forests of the western United States (US), observa-
tions of limited postfire conifer regeneration have led to the hypoth-
esis that forested stands may be replaced by persistent grasslands or
shrublands following fire, at least within portions of burns in
which seed availability is low (Dodson and Root 2013;Keyser et al.
2008;Roccaforte et al. 2012;Savage and Mast 2005). In the Colo-
rado Front Range (CFR) and throughout the western US, more
research is needed to document whether current patterns of post-
fire conifer regeneration are incongruous with historic patterns
and what factors (e.g., increased temperatures and associated
drought, wildfire severity, etc.) explain any significant deviation
from the past. We tackle part of this complicated issue through a
field experiment that assesses the role that variability in temper-
ature and water plays in influencing the growth rates and percent
survival of ponderosa pine (Pinus ponderosa Douglas ex P. Lawson &
C. Lawson) and Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco)
seedlings that were planted in a recently disturbed environment.
Our research will allow land managers to better understand and
prepare for changes in patterns of postfire conifer regeneration,
including potential shifts from forest to nonforest vegetation.
In Colorado, temperatures have risen almost universally across
the state in recent decades and are expected to increase by an addi-
Received 22 January 2015. Accepted 21 July 2015.
M.T. Rother, T.T. Veblen, and L.G. Furman. Biogeography lab, Department of Geography, University of Colorado, Campus Box 260, Boulder,
CO 80309, USA.
Corresponding author: Monica T. Rother (e-mail: rother@colorado.edu).
1607
Can. J. For. Res. 45: 1607–1616 (2015) dx.doi.org/10.1139/cjfr-2015-0033 Published at www.nrcresearchpress.com/cjfr on 30 September 2015.
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF COLORADO on 10/04/15
For personal use only.
tional 1.4–3.6 °C by 2050 (Lukas et al. 2014;Reclamation 2013). Unlike
predictions of changing temperature, uncertainty surrounds how
precipitation regimes might change, and over the last several
decades, there has been no clear trend across the state (Lukas et al.
2014;Reclamation 2013). In the absence of increased precipitation,
higher temperatures are associated with an increased occurrence
of drought due to higher rates of evapotranspiration. Increased
drought has already resulted in ecological change in many forested
communities including higher background tree mortality rates (van
Mantgem et al. 2009;Williams et al. 2013). Researchers have linked
tree mortality to carbon starvation, hydraulic failure, or a combina-
tion of the two, along with other drought-mediated factors such as
insect attack and disease (McDowell et al. 2011;Sevanto et al. 2014).
Trees that do survive climate stress may acclimate in a variety of
ways such as through biomass partitioning (Turner 1997). A number
of studies have shown that conifers allocate more carbon to roots
and (or) sapwood under conditions of elevated temperature, water
stress, or both (Callaway et al. 1994;Delucia et al. 2000;Olszyk et al.
2003). Although this strategy may improve the likelihood of survival
by increasing water access, it may also increase the probability of
death by limiting plant height and increasing susceptibility to mor-
tality by factors such as wildfire, herbivory, and competition for
light.
Land managers and researchers have long recognized the impor-
tance of climate variability in driving patterns of conifer regenera-
tion in dry ponderosa pine forests. For example, many early papers
identified 1919 as an astonishing year for widespread ponderosa
pine regeneration in Arizona, due to high levels of summer pre-
cipitation that followed an excellent seed year (Cooper 1960;
Pearson 1923). This observational evidence of infrequent years of
abundant regeneration coinciding with favorable weather was
supported in later years through the development of large data-
sets of annually resolved establishment dates in a ponderosa pine
forest in Arizona (Savage et al. 1996) and along a grassland–forest
ecotone of the CFR (League and Veblen 2006). These studies both
concluded that in environments lacking recent disturbance, es-
tablishment by ponderosa pine occurs episodically in association
with high moisture availability. More recently, researchers exam-
ined relationships between climate variability and ponderosa
pine regeneration following wildfire (Feddema et al. 2013;Savage
et al. 2013) and found that monthly to seasonal climate conditions
associated with multiple developmental stages of ponderosa pine
(e.g., cone production, germination, etc.) were important for predict-
ing patterns of observed postfire ponderosa pine regeneration. Fur-
ther research is needed to document whether similar relationships
between climate variability and postfire ponderosa pine regenera-
tion hold true in the CFR, given significant differences between the
two areas (e.g., climate regimes, soil characteristics, understory com-
position, genetic provenance of species, etc.).
In addition to climate conditions, fire severity may also influ-
ence patterns of conifer regeneration in dry ponderosa pine for-
ests. Within a given burn, large patches of high-severity fire can
limit regeneration by ponderosa pine and Douglas-fir because these
species disperse seed primarily by wind over relatively short dis-
tances of c. 200 m or less (Bonnet et al. 2005;Haire and McGarigal
2010;Shatford et al. 2007). Additionally, patches of high-severity
wildfire can create altered microclimate conditions such as higher
daily temperature ranges and reduced soil moisture levels due to
blackened soil and absence of vegetation (Montes-Helu et al. 2009;
Ulery and Graham 1993). However, abundant conifer regeneration
following high-severity fire has been documented in ponderosa pine
forests (Ehle and Baker 2003;Haire and McGarigal 2010;Savage and
Mast 2005;Veblen and Lorenz 1986), indicating that high-severity fire
does not universally result in regeneration failure. Additionally, in
the ponderosa pine zone in the CFR, the historic wildfire regime was
mixed severity, meaning that fire effects were varied both within
stands and across the landscape and included low-, moderate- and
high-severity fire (Sherriff and Veblen 2007). Fire severity undoubt-
edly plays an important role in influencing postfire vegetation tra-
jectories in dry ponderosa pine forests, but it is unlikely to be the sole
factor explaining observations of limited conifer regeneration fol-
lowing recent wildfires.
In the present study, we focused on the role that differences in air
temperature and water availability play in influencing postdistur-
bance conifer regeneration in low-elevation forests of the CFR. We
employed open-top chambers and watering treatments to assess
how altered temperature and water availability influenced growth
rates and percent survival of ponderosa pine and Douglas-fir seed-
lings at a site where the aerial biomass was scraped off to expose bare
mineral soil, simulating fire (i.e., “scalping” sensu Kayes et al. (2010)).
Our primary objectives were to (i) examine the effects of manipula-
tions of temperature and water on the growth rates and percent
survival of conifer seedlings, (ii) assess potential differences in
aboveground vs. belowground biomass partitioning by conifer seed-
lings, and (iii) determine whether conifer seedling growth and sur-
vival were dependent on herbaceous and shrub groundcover. We
hypothesized that experimental treatments would result in signifi-
cant differences in growth and survival patterns of both ponderosa
pine and Douglas-fir. Given the semi-arid, low-elevation setting for
the experiment, we expected that increased air temperature would
result in decreased growth rates and percent survival, whereas in-
creased water would result in increased growth rates and percent
survival. We hypothesized that ponderosa pine growth rates and
percent survival may be higher than Douglas-fir growth rates and
percent survival given that the latter species tends to occupy rela-
tively cooler and more mesic sites, although significant overlap of
the two species occurs. Partitioning among aboveground vs. below-
ground biomass of the conifer seedlings was also expected to vary;
we hypothesized that allocation to belowground biomass, as op-
posed to aboveground biomass, would be greater under conditions
of higher water stress. Finally, nonconifer biomass was expected to
vary in response to the experimental treatments; high total nonco-
nifer biomass was expected to be associated with lower growth rates
and percent survival of conifer seedlings due to increased competi-
tion for water.
Materials and methods
Study area
We installed the experiment on a closed section of Heil Valley
Ranch Open Space in Boulder County, Colorado. The research site
was located at 40.15°, −105.32° at an elevation of 1960 m within the
lower montane zone of the CFR. The experimental plot was situated
on a slope with a north–northeast aspect and a gentle gradient,
surrounded by dry ponderosa pine forest. Several juvenile pon-
derosa pine trees were present in the plot prior to experimentation,
indicating that the site was suitable for conifer regeneration. Data
from a nearby weather station (Boulder Station, 1893–2013, Western
Regional Climate Center, 39.99°, −105.27°) show that the mean max-
imum January temperature for the area is c. 7.2 °C and mean maxi-
mum July temperature is c. 30.2 °C. Total annual precipitation is c.
476 mm. Precipitation patterns vary significantly over the course of
the year, as well as interannually. In typical years, peak precipitation
occurs in spring and summer months.
Experimental design
Preparation for the field experiment began in the spring of
2012. A macro plot of 35m×42mwasdivided into a grid of
120 cells of 3.5m×3.5m(Fig. 1). Due to excessively rocky or uneven
surfaces in portions of the macro plot, only 100 of the cells were
selected for use in the experiment. In each of the 100 cells, burn-
ing was simulated by killing all aerial biomass through scraping
away of the vegetation to expose bare mineral soil (i.e., “scalping”
sensu Kayes et al. (2010)). Prescribed fire was not a viable option
because a statewide burn ban was in place at the time. Although the
effects of scalping are not identical to those of burning, an advantage
1608 Can. J. For. Res. Vol. 45, 2015
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF COLORADO on 10/04/15
For personal use only.
of this approach for field experiments is that it creates a more uni-
form disturbance than the patchier effect of prescribed fire. After
scalping was completed, we obtained 700 ponderosa pine and 700
Douglas-fir seedlings from the Colorado State Forest Service seedling
tree program. We then planted fourteen seedlings of ponderosa pine
and Douglas-fir (seven seedlings each) in a hexagonally shaped area
of 2.6 m
2
within each cell, hereafter termed the micro plot. The
seedlings were from a local genetic provenance and were approxi-
mately 1 year old at the time of planting. We regularly watered all of
the seedlings for approximately 1 month prior to initiating experi-
mental manipulations to help them acclimatize and minimize initial
mortality. A fence that was c.1.8 m tall was installed around the
perimeter of the macro plot to deter entry and subsequent trampling
and (or) herbivory by large mammals such as elk and deer. Following
site preparation and the acclimation period, one of the following
four treatments was assigned randomly to each micro plot: (i)
warmed only (Wm), (ii) watered only (Wt), (iii) warmed and watered
(WmWt), and (iv) control (Co). Treatment types were assigned follow-
ing a restricted random protocol to ensure that minor variability in
soil characteristics, slope, ground cover, and light availability within
the macro plot did not confound the study's outcomes.
For micro plots designated to be warmed (Wm and WmWt), hex-
agonal open-top chambers (OTCs) were constructed and installed
following the methods outlined by Marion et al. (1997). The OTCs
were expected to increase air temperature by c. 1–2 °C (Hollister and
Webber 2000;Marion et al. 1997;Tercero-Bucardo et al. 2007), which
is reasonable given the 1.4–3.6 °C forecasted temperature increase
expected by 2050 in Colorado (Lukas et al. 2014;Reclamation 2013).
More uncertainty surrounds how precipitation regimes might
change in Colorado (Lukas et al. 2014;Reclamation 2013), and over
the last several decades, there has been no clear trend (Lukas et al.
2014). In this study, we simulated increased precipitation through
watering treatments. For micro plots designated to be watered (Wt
and WmWt), weekly watering treatments were implemented to
roughly approximate the upper quartile for total monthly precipita-
tion, based on local instrumental climate data (Boulder Station,
1893–2010, Western Regional Climate Center). To do so, the differ-
ence between the long-term median and upper quartile for monthly
total precipitation was first calculated. Then, we determined what
additional volume of water would be required to raise the median
value to the upper quartile, based on the surface area to be watered.
This calculation led to varying weekly watering requirements for
each month (June, 9.8 L; July, 9.1 L; August, 6.1 L; September, 9.8 L).
However, for feasibility purposes related to water delivery to the site
and the manual implementation of the watering treatments, we
used a consistent watering treatment of 7.6 L·week
−1
per micro plot.
This method does not fully replicate precipitation, as natural precip-
itation falls more variably in terms of both the amount of water in a
single rainfall event and the time between rainfall events. However,
we assumed that our watering efforts would effectively alter mois-
ture availability in similar ways as natural rainfall (i.e., through in-
creased soil moisture). Both warming and watering treatments
occurred only during the growing season (June–September). During
excessively dry periods, additional watering was occasionally deliv-
ered to all 100 micro plots to compensate for excessive drought that
could lead to widespread mortality of seedlings across all treatment
types. It is important to note that growing season precipitation pat-
terns in the study area are highly variable and that excessively dry
periods are not uncommon. Thus supplemental watering during
those periods may have resulted in higher growth rates and percent
survival across all treatment types than would otherwise be ex-
pected.
Data collection
We used HOBO automated data loggers (Onset Computer Corp.,
Bourne, Massachusetts) to monitor how the experimental treat-
ments influenced air temperature, relative humidity, and soil
temperature. We used restricted random methods to identify
eight micro plots (two of each treatment type) for monitoring. In
each of these micro plots, we placed (i) one data logger for air
temperature and relative humidity, situated 20 cm aboveground,
near the seedling canopy, and (ii) two data loggers for soil temper-
ature, buried at a depth of 5 cm. Air temperature and relative
humidity data were recorded every 30 min, whereas soil temper-
ature data were recorded every 20 min (the highest allowable
frequency given the data storage limitations of the devices). We
monitored changes in ponderosa pine and Douglas-fir seedling
growth rates and percent survival over the 2-year period. Data
collection of seedling stem height and status as dead or living
occurred at the start and end of both growing seasons (2012 and
2013), for a total of four data collection periods. A seedling was
considered dead if no green needles remained. At the end of the
Fig. 1. Plot design for the study including (A) site location within the Colorado Front Range, (B) layout of randomly located experimental
treatments within the macro plot (1, warmed only (Wm); 2, warmed and watered (WmWt); 3, watered only (Wt); 4, control (Co)), (C) cell, and
the (D) micro plot within the cell in which experimental treatment was applied.
A = 2.6 m
2
Denver
Site location
Colorado
(A)
(B)
(C)
(D)
35 m
42 m
3.5 m
1 m
1 3 3 4 4 4 1 1 4 2
2 2 1 1 4 2 3 3 4
3 4 2 3 3 2 4 3 1 1
1 1 1 4 2 3 3 4 2 2
3 3 1 2 1 3 4 4 3
1 4 2 2 1 1 4 4 2 3
1 3 3 2 2 1 4 2 1
4 4 2 3 4 3 1
1 2 2 2 3 3 4
4 3 1 3 2 1 2 3
4 1 2 4 4 3 2
1 4 2 1
!
Rother et al. 1609
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF COLORADO on 10/04/15
For personal use only.
second growing season (i.e., 2013), we harvested all biomass in a
subset of the micro plots (n= 80) to calculate the aboveground and
belowground biomass of conifer seedlings, as well as the aboveg-
round herbaceous and shrub biomass (hereafter, “nonconifer bio-
mass”). The nonconifer biomass data were collected to determine
whether experimental treatments influenced nonconifer biomass
and to assess whether competitive effects were important to co-
nifer seedling growth rates and percent survival. All harvested
biomass samples were weighed after they were dried in an oven at
70 °C to remove all water mass.
Data analyses
We analyzed the data from the data loggers to determine how
the experimental treatments influenced air temperature, relative
humidity, and soil temperature. We calculated mean air temper-
ature, relative humidity, and soil temperature for each time step
(e.g., 12:00 AM, 12:30 AM, 1:00 AM) over the course of a day for the
full experimental period (June–September) in both 2012 and 2013.
We also assessed ambient conditions in 2012 and 2013 using local
climate data (Boulder Station, 1893–2011, Western Regional Cli-
mate Center) to compare conditions in those years with the long-
term mean. To assess whether experimental treatments affected
conifer seedling growth rates and percent survival, we pooled
individual tree seedling stem height and survivorship data to the
micro plot level and calculated mean height growth rates and
percent survival for the 2012 and 2013 growing seasons. Height
growth rates were calculated as the mean percent increase in
seedling stem height that occurred from the start to the end of
each growing season, averaged for each micro plot, for each spe-
cies. Percent survival was calculated as the percentage of seed-
lings of each species in a micro plot that survived from the start to
the end of each growing season. Height growth rates and percent
survival for each of the four treatment types were then assessed
using generalized linear models (GLMs) with robust standard er-
rors in Stata software (StataCorp 2015). Robust standard errors
were used to account for clustering in the data due to noninde-
pendence of plots between years (Baum 2006). With regard to our
height growth rate data, log transformations were applied to data
for Douglas-fir to address issues of non-normality. In the case of
the percent survival data, the GLMs applied a logit link function
with a binomial family specified. We also assessed differences in
ratios of root to shoot biomass, as well as nonconifer biomass, by
treatment type, again using data pooled to the micro plot level,
using 2×2 ANOVA.
Results
Temperature and relative humidity data
We observed that air temperature, relative humidity, and soil
temperature varied among treatment types, particularly at mid-
day (10:00 AM to 6:00 PM). Mean midday air temperature in Wm
plots was 4.4 °C warmer than in Co plots in 2012 and 3.2 °C warmer
than in Co plots in 2013 (Fig. 2;Table 1). In contrast, differences in
mean air temperatures among the treatment types for the entire
24-hour day were less pronounced; we observed a 1.63 °C differ-
ence between Wm and Co in 2012 and a 1.38 °C difference between
Wm and Co in 2013. With regard to relative humidity, results were
consistent with expectations that over the course of a day, relative
humidity would typically be lowest at midday, when temperatures
were highest (Fig. 2). Among treatment types, mean midday relative
humidity was lowest in Wm plots in both years. Additionally, we
observed that both mean midday air temperature and mean midday
relative humidity were more variable in Wm and WmWt plots than
in Wt and Co plots in both years, as indicated by the relatively high
standard deviations (SDs) associated with those treatment types. Fi-
nally, we observed that differences in soil temperature were less
pronounced than differences in air temperature and relative humid-
ity, although the Wm plots were associated with higher soil temper-
atures.
Ambient climate conditions
In both 2012 and 2013, there were periods during which the
monthly air temperature or precipitation deviated substantially
from the long-term mean (Table 2). Regarding air temperature, all
but one month during the experimental period (July 2013) was
hotter than the long-term mean (1893–2012). Months in which
average temperatures were especially high included June 2012,
September 2012, and June 2013. Mean temperatures during those
months exceeded the long-term mean by more than 1 SD. Addi-
tionally, 2012 and 2013 were notably different from each other,
with higher mean monthly temperatures occurring in 2012 than
in 2013. This general pattern of higher temperatures in 2012 than
in 2013 documented by the climate station data (Table 2) is also
evident in the air temperature data from the field experiment
(Fig. 2;Table 1). Mean and median air temperatures in all four
treatment types were warmer in 2012 than in 2013. With regard to
precipitation, some monthly totals during the experimental pe-
riod were exceptionally low, whereas others were exceptionally
high. June 2012, August 2012, and June 2013 had total precipitation
amounts that were more than 1 SD lower than the long-term mean
(Table 2). These were months during which supplemental water
was applied to the entire experimental plot to compensate for lack
of natural rainfall. In terms of wet periods, July 2012 and Septem-
ber 2013 had total precipitation amounts that were more than
1 SD above the long-term mean. September 2013 is especially
remarkable as during part of that month, flood conditions oc-
curred in Boulder and across a broad stretch of the CFR. Total
monthly precipitation was 46.1 cm (Boulder Station, Western Re-
gional Climate Center), which was approximately 13 SDs above
the long-term mean. The experiment was terminated approxi-
mately 2 weeks after this event, as soon as access was feasible.
Although the study site received an enormous amount of precip-
itation in September, no significant changes were observed (e.g.,
no major erosion, no damage to OTCs or monitoring equipment,
and no significant changes in seedling survival or growth).
Ponderosa pine growth and survival
In both 2012 and 2013, we observed that measurements of
growth and survival for ponderosa pine seedlings varied among
treatment types (Fig. 3). Generally, growth rates and percent sur-
vival were highest in the Wt plots and lower in the Co, WmWt,
and Wm plots, in that order. Differences between Wm and Wt
were especially large. In 2012, the median height growth rate was
5.6% in Wm plots compared with 15.5% in Wt plots, and in 2013,
median height growth rates were 6.2% and 11.7% for Wm and Wt
plots, respectively. Our GLMs revealed numerous significant rela-
tionships (Table 3). In the case of our ponderosa pine height growth
rate GLM, our findings indicate that the Wm treatment resulted
in significantly lower height growth rates compared with the
control (P< 0.001) and the Wt treatment resulted in significantly
higher height growth rates compared with the control (P< 0.05).
Regarding the survival of ponderosa pine seedlings, median per-
cent survival in Wm plots in 2012 and 2013 were 42.9% and 66.7%,
respectively, whereas in both years, the median percent survival
in Wt plots was 100%. With regard to the ponderosa pine survival
GLM, we found significantly lower odds of survival for plots with
the Wm and WmWt treatment (P< 0.001) and significantly higher
odds of survival for plots with the Wt treatment (P≤ 0.01). In terms
of year of experiment, we found significantly lower ponderosa
pine height growth rates in 2013 vs. 2012 (P< 0.001) and signifi-
cantly higher odds of ponderosa pine percent survival in 2013 vs.
2012 (P< 0.001). Finally, results from 2x2 ANOVA indicate that root
to shoot ratios differed by treatment type (F= 11.98, P< 0.001). Root
to shoot ratios were higher in Wm and WmWt plots than in Wt
and Co plots (Fig. 4;Table 4).
1610 Can. J. For. Res. Vol. 45, 2015
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF COLORADO on 10/04/15
For personal use only.
Douglas-fir growth and survival
We observed that Douglas-fir growth rates and percent survival
were typically highest in the Wt plots followed by the Co, WmWt,
and Wm plots, respectively (Fig. 3). This general pattern is the
same as what we observed for ponderosa pine. However, for
Douglas-fir, both height growth rates and percent survival were
lower than for ponderosa pine regardless of treatment type. In
2012, for all treatment types, the median height growth rate of
Douglas-fir seedlings was less than 4%; this is even lower than the
lowest median height growth rate for ponderosa pine that year
(5.6% in Wm plots). The Douglas-fir height growth rate GLM indi-
cated that the watering treatment resulted in significantly higher
height growth rates compared with control (P< 0.05). In terms of
survival of Douglas-fir seedlings, percent survival in both 2012 and
2013 were lowest in Wm plots, with medians of 28.6% and 16.7%,
respectively (Fig. 3). In contrast, Wt plots had median percent
Fig. 2. Mean air temperature, relative humidity, and soil temperature by treatment type for (A) year 1 (2012) and (B) year 2 (2013)
of the experiment. Data were collected for June–September using HOBO data loggers placed in a subset of the plots. Different line types
designate different treatment types: solid lines, warmed only (Wm), long-dashed lines, warmed and watered (WmWt), short-dashed lines,
watered only (Wt), and dotted lines, control (Co).
15 20 25 30 35
20 4030 6050 80
70
Mean Relative Humidity (%)
(A) 2012 (B) 2013
15 20 25 30 35
12am 4am 8am 12pm 4pm 8pm 12am 12am 4am 8am 12pm 4pm 8pm 12am
Time of Day
Table 1. Midday (10 AM – 6 PM) air temperature, relative humidity, and soil temperature by treatment
type during the experimental period (June–September of 2012 and 2013).
Air temperature (°C) Relative humidity (%) Soil temperature (°C)
Mean Median SD Mean Median SD Mean Median SD
2012
Wm 33.2 34.6 4.2 27.4 25.8 5.4 30.7 31.4 3.0
WmWt 32.6 33.6 3.8 29.7 28.4 5.3 29.4 30.3 3.0
Wt 29.3 30.1 2.7 31.5 30.4 4.4 28.5 29.2 3.1
Co 28.8 29.5 2.5 31.6 30.6 4.1 29.6 30.3 2.8
2013
Wm 30.4 31.2 4.0 41.1 39.9 6.3 29.9 30.4 2.7
WmWt 28.8 29.7 3.2 45.4 43.6 6.0 26.3 26.7 1.7
Wt 27.3 28.0 2.7 45.0 43.5 4.7 26.9 27.3 2.1
Co 27.2 27.9 2.8 46.5 45.0 5.1 27.5 27.9 2.2
Note: SD, standard deviation. Treatment types: Wm, warmed only; WmWt, warmed and watered; Wt, watered
only; Co, control.
Rother et al. 1611
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF COLORADO on 10/04/15
For personal use only.
survivals of 85.7% and 100% for 2012 and 2013, respectively. Our
Douglas-fir survival model revealed significantly lower odds of
survival for plots with the Wm (P< 0.001) and WmWt (P< 0.05)
treatments. With regard to year of experiment, we found signifi-
cantly higher Douglas-fir height growth rates in 2013 vs. 2012
(P< 0.001) and significantly higher Douglas-fir percent survival in
2013 vs. 2012 (P< 0.001). Lastly, in terms of aboveground and
belowground biomass, results from 2×2 ANOVA indicate that root
to shoot ratios varied by treatment type (F= 16.17, P< 0.001). Root
to shoot ratios were higher in Wm and WmWt plots than in Wt
and Co plots.
Nonconifer biomass
By the end of the second experimental year (i.e., 2013), most
plots had substantial cover by nonconifer biomass, mostly by grasses
and forbs. Among all 100 micro plots, total nonconifer biomass var-
ied widely from 12.9 g·m
−2
to 98.3 g·m
−2
, with an average of 43.3 ±
17.8 g·m
−2
(mean ± SD). Some plots still had substantial amounts of
bare soil at the end of 2013, whereas other plots were completely
vegetated. Although nonconifer biomass varied substantially across
the entire macro plot, no statistically significant relationships were
found through our 2×2 ANOVA (Fig. 5), indicating that the experi-
mental treatments did not explain the observed differences.
Discussion
Our study provides significant insight regarding the effects of
climate change on post-disturbance vegetation patterns. Our find-
ings suggest that warming temperatures and associated drought
are likely to inhibit postfire regeneration of ponderosa pine and
Douglas-fir in low-elevation forests of the CFR and that future
postfire vegetation composition and structure may differ notably
from historic patterns in some areas. Because we planted tree
seedlings, our experiment focuses specifically on the growth rates
and percent survival of conifer seedlings in the absence of seed
limitation. Masting by ponderosa pine in the CFR is known to be
sensitive to climate (Mooney et al. 2011), and thus research is
needed to investigate how the changing climate may affect seed
production. The experimental treatment in our study that is most
similar to what is expected for the future in the study area is the
Wm treatment, in which temperatures are elevated but the
amount of precipitation does not significantly change. Height
growth rates and percent survival for both ponderosa pine and
Douglas-fir seedlings in Wm plots were much lower than in other
treatment types. Although we think that the Wm scenario most
closely resembles future climate conditions based on model pro-
jections (Lukas et al. 2014;Reclamation 2013), there is consider-
able uncertainty about how precipitation regimes may change in
the CFR. It is possible that elevated temperatures will be accom-
panied by increased precipitation (like the WmWt treatment) or
reduced precipitation (not examined in this study). However, even
if precipitation is to increase in the future, our findings indicate
that ponderosa pine and Douglas-fir growth rates and percent
survival may still be relatively limited, as indicated by the differ-
ence between WmWt and Co treatment types. Changes in pat-
terns of postfire conifer regeneration in lower montane forests of
the CFR have important management implications given the eco-
logical, social, and economic benefits these forests provide (e.g.,
carbon storage, habitat, recreation, timber, etc.).
Effects of experimental treatments on temperature and
relative humidity
The experimental treatments we implemented effectively al-
tered conditions in the plots. With regard to air temperature, the
average increased temperature we achieved was similar to the c.
1–2 °C documented in other studies that used OTCs (Hollister and
Webber 2000;Marion et al. 1997;Tercero-Bucardo et al. 2007) and is
reasonable given expectations of an increase of 1.4–3.6 °C in Colo-
rado in future years (Lukas et al. 2014;Reclamation 2013). Also similar
to other studies, temperature differences between experimental
treatments were most pronounced during midday (Marion et al.
1997;Tercero-Bucardo et al. 2007). The similarity in air tempera-
ture among experimental treatments outside of the midday hours
indicates that the OTCs did not create uniform warming through
time but, instead, resulted in amplified temperatures only when
incoming solar radiation was present. Trapped heat was largely or
completely lost from the OTCs at night. In contrast, warming
conditions associated with human-induced climate change result
in elevated temperatures during both daytime and nighttime.
This is a notable limitation given that warmer nighttime temper-
atures may have significant effects on plant growth rates and
percent survival (Turnball et al. 2002).
In addition to air temperature differences, we also observed
differences in relative humidity among treatment types. Findings
indicated that warmed plots had drier air than plots that did not
receive a warming treatment. Soil moisture was not monitored,
but we assume that soil moisture varied by treatment type be-
cause height growth rates and percent survival of ponderosa pine
and Douglas-fir seedlings were higher in Wt plots than in Co plots.
Regarding soil temperatures, we observed relatively small differ-
ences in soil temperatures among treatment types, suggesting
that air temperature, relative humidity, and soil moisture were
more important than soil temperature in driving differences in
the growth rates and percent survival of ponderosa pine and
Douglas-fir seedlings.
Growth and survival of ponderosa pine and Douglas-fir
seedlings
Our experiment was situated in the lower montane zone of the
CFR, where moisture limits tree growth. As hypothesized, we found
that increased temperatures generally resulted in lower average per-
cent survival and height growth rates for both ponderosa pine and
Douglas-fir seedlings, whereas increased water resulted in higher
average percent survival and height growth rates for both ponderosa
pine and Douglas-fir seedlings. We interpret lower percent survival
and height growth rates in warmed plots (Wm and WmWt) to be a
consequence of lower moisture availability in those treatment types,
due to increased evapotranspiration. However, we did not monitor
plant physiological responses and thus cannot be certain what mech-
anisms drove the plant responses we observed; further study to that
end would be beneficial. With regard to root to shoot ratios, we
found that the warming treatment (Wm and WmWt) had a signifi-
cant effect on mean root to shoot ratios, with higher root to shoot
ratios in plots that were warmed. Previous studies of conifer biomass
partitioning (Callaway et al. 1994;Delucia et al. 2000;Olszyk et al.
2003) indicate that water-stressed conifers tend to allocate more car-
Table 2. Climate conditions during the exper-
imental period (2012 and 2013) vs. the long-
term record.
Long-term
record
2012 2013 Mean SD
Mean temperature (°C)
June 23.4 21.1 19.3 1.7
July 23.8 22.3 22.5 1.5
August 22.9 22.3 21.7 2.3
September 18.9 18.4 17.2 1.3
Total precipitation (cm)
June 1.0 1.5 4.9 3.4
July 12.7 2.6 4.7 3.0
August 0.9 3.6 4.1 3.0
September 5.8 46.1 3.9 3.2
Note: Data are from the Boulder Station, 1893–2011,
Western Regional Climate Center. SD, standard devi-
ation.
1612 Can. J. For. Res. Vol. 45, 2015
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF COLORADO on 10/04/15
For personal use only.
bon to their roots to increase their ability to access soil water. This
strategy may reduce the probability of mortality due to water stress
but increase the likelihood of death by drivers where height is ben-
eficial (e.g., wildfire, herbivory, and competition for light). Although
many trends were similar for ponderosa pine and Douglas-fir, we
observed that growth rates and percent survival for Douglas-fir was
universally lower than for ponderosa pine. This finding suggests
that Douglas-fir seedlings were more stressed than ponderosa
pine seedlings. Both ponderosa pine and Douglas-fir can tolerate
dry conditions, but where the two species co-occur, ponderosa
pine is more common in relatively xeric topographic settings (i.e.,
low-elevation, south-facing aspects) compared with Douglas-fir
(Kaufmann et al. 2006;Peet 1981), and our findings are also con-
sistent with experimental work that demonstrated that Douglas-
fir may be more vulnerable to drought stress than ponderosa pine
(Cleary 1970).
Nonconifer biomass
We did not observe significant differences in nonconifer bio-
mass among treatment types, suggesting that altered tempera-
ture and water did not influence the growth rates and percent
survival of forbs, grasses, and shrubs. This indicates that these
understory plants were less sensitive to the experimental treat-
ments we implemented than the ponderosa pine and Douglas-fir
seedlings and thus may be less affected by future climate change in
the lower montane zone. Although there was significant variability
in total nonconifer biomass among the micro plots, this variability
did not correspond to differences in seedling growth rates and per-
cent survival. We interpret this finding to indicate that competitive
effects between nonconifer plant species and ponderosa pine and
Douglas-fir seedlings were not significant in our study. The differ-
ences in temperature and relative humidity created by the experi-
mental treatments, not variability in nonconifer biomass, was the
Fig. 3. Mean height growth rate (%) and survival (%) by treatment type for ponderosa pine and Douglas-fir seedlings for (A) year 1 (2012) and (B)
year 2 (2013) of the experiment. The thick black line inside the box indicates the median, the lines at the outer edges of the box indicate the
upper and lower quartiles, and the lines at the end of vertical dashed lines indicate the maximum and minimum values. The dots indicate any
outliers. Treatment types: Wm, warmed only; WmWt, warmed and watered; Wt, watered only; Co, control.
0 5 10 15 20 25 30
Height Growth Rate (%)
Wm WmWt Wt Co
0 20406080100
Survival (%)
Wm WmWt Wt Co Wm WmWt Wt Co Wm WmWt Wt Co
(A) 2012 (B) 2013
Ponderosa pine Douglas-fir Douglas-fir Ponderosa pine
Treatment Type Treatment Type
Table 3. Generalized linear models (GLMs) of height growth rate and
percent survival for ponderosa pine and Douglas-fir seedlings.
Coefficient (SE) Odds ratio (SE)
Ponderosa
pine growth
model
ln Douglas-fir
growth
model
Ponderosa
pine survival
model
Douglas-fir
survival
model
Wm −4.78 (0.95)*** −0.11 (0.18) 0.16 (0.04)*** 0.16 (0.05)***
Wt 2.41 (0.97)* 0.41 (0.17)* 4.69 (2.10)** 1.77 (0.52)
Wm × Wt −1.71 (1.11) 0.15 (0.17) 0.31 (0.09)*** 0.49 (0.14)*
Year −3.04 (0.62)*** 0.81 (0.16)*** 2.74 (0.49)*** 2.09 (0.42)***
Constant 12.87 (0.87) 0.70 (0.13) 5.49 (1.27) 2.25 (0.52)
Note: Models are based on the combined dataset for both years. Standard
errors (SE) are robust standard errors. Asterisks indicate statistical significance:
*, P< 0.05; **, P< 0.01; ***, P< 0.001. Treatment types: Wm, warmed only;
Wt, watered only.
Fig. 4. Shoot and root biomass (g·m
−2
) by treatment type for (A)
ponderosa pine and (B) Douglas-fir seedlings. Biomass harvesting
was completed at the end of year 2 (2013) of the experiment.
Treatment types: Wm, warmed only; WmWt, warmed and watered;
Wt, watered only; Co, control.
Shoot Biomass (g/m2)
01020
Root Biomass (g/m2)
Wm WmWt Wt Co
10 0
Wm WmWt Wt Co
Treatment Type
(A) Ponderosa pine (B) Douglas-fir
Rother et al. 1613
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF COLORADO on 10/04/15
For personal use only.
key driver of patterns of ponderosa pine and Douglas-fir growth rates
and percent survival.
Forest management in the context of climate change
Disturbances such as wildfire can act as catalysts of rapid trans-
formation of forested ecosystems, particularly under changing
climate conditions. Although mature trees often survive through
suboptimal environments such as temperatures outside the pre-
ferred range of the species, new germination and establishment
depends on a relatively narrow range of requirements and can be
highly responsive to subtle changes in climate (Hogg and Schwarz
1997;Spittlehouse and Stewart 2003). A general hypothesis of
rapid post-disturbance shifts in vegetation patterns in the context
of changing climate has previously been put forward by numer-
ous researchers (e.g., Enright et al. 2015;Hogg and Schwarz 1997;
Johnstone et al. 2010a;Spittlehouse and Stewart 2003;Turner
2010), yet few studies have tested this expectation (but see Dodson
and Root 2013;Feddema et al. 2013;Hogg and Schwarz 1997;
Johnstone et al. 2010b;Landhausser and Wein 1993;Moser et al.
2010;Overpeck et al. 1990;Savage et al. 2013;Tercero-Bucardo
et al. 2007). Our field experiment directly assessed how tempera-
ture and water availability influence tree seedling growth rates
and percent survival after disturbance and contributes to the
growing understanding that in the context of climate change,
disturbances such as wildfires may result in vegetation patterns
inconsistent with predisturbance patterns.
We demonstrated that ponderosa pine and Douglas-fir seedling
growth rates and percent survival following disturbance was
inhibited by warmer temperatures but that nonconifer biomass
(i.e., grasses, forbs, and some shrubs) was unaffected by the exper-
imental treatments. Historically, conifer regeneration in dry pon-
derosa pine forests nearby and in the CFR was abundant after fire,
as indicated by tree-ring retrospective studies that document pon-
derosa pine and Douglas-fir establishment following wildfires in
the 19th century and earlier (Ehle and Baker 2003;Mast et al. 1998;
Veblen and Lorenz 1986). Our findings suggest that future postfire
vegetation trajectories in lower montane ponderosa pine forests
of the CFR may differ notably from historic patterns. In the ab-
sence of abundant regeneration by ponderosa pine and Douglas-
fir, some previously forested areas may be replaced by persistent
grasslands or shrublands, particularly at lower elevations near the
ecotone, where water stress is highest. Where conifer regenera-
tion does occur, our findings suggest that ponderosa pine may be
more common than Douglas-fir. Given that our study does not
account for seed limitation (removed by planting seedlings) or
periods of extreme drought (removed by supplemental watering
in periods without rainfall), our findings are conservative and
may underestimate the consequences of warmed climate on pon-
derosa pine and Douglas-fir regeneration. Land managers in the
CFR should be prepared for postfire vegetation trajectories that
are incongruent with historic patterns. Difficult decisions will be
required concerning how to manage recently burned, lower mon-
tane forests given expectations of less abundant regeneration by
ponderosa pine and Douglas-fir. For high-priority areas, land man-
agers may choose to adopt resistance strategies (Millar et al. 2007)
to forestall major vegetation changes. Given that wildfire has the
potential to drive rapid vegetation change, fire mitigation prac-
tices such as installing fuel breaks, thinning stands, or using pre-
scribed burning may be appropriate. However, management
strategies such as these can be expensive, time intensive, and of
variable efficacy (Graham et al. 2012;Roccaforte et al. 2010) and
are, therefore, only viable at relatively small spatial scales. Land
managers may also opt for strategies that promote resiliency of
forests to wildfire. Resilient forests are described as those that
return to a similar prior condition after disturbance (Holling 1973;
Millar et al. 2007). Intensive management of the postfire landscape
to promote resiliency may include tree plantings. Our study suggests
that plantings should occur on cooler, wetter microsites such as
north-facing aspects. Additionally, plantings that are timed around
cooler, wetter periods (such as during El Niño conditions in the CFR)
are likely to be most successful. Ultimately, response strategies that
accept the transition of postfire landscapes to new vegetation assem-
blages may be most feasible and practical at broad scales. Such
largely passive response strategies may be ideal for remote areas
where access is limited and costs of active management are high.
One of the most significant impacts of localized climate-induced
transitions from forest to nonforest vegetation is likely to be on
water quantity and quality, which should be an area of priority re-
search (Ebel and Mirus 2014).
In conclusion, the results of our study suggest that under pro-
jected climate change scenarios, post-disturbance regeneration of
ponderosa pine and Douglas-fir may be limited in lower montane
forests of the CFR. Different species assemblages are expected to
emerge as regeneration occurs most abundantly among the spe-
cies best suited to the current climate, rather than those favored
previously. In recently burned areas of high tree mortality, conifer
Table 4. Ratios of root to shoot biomass
by treatment type.
Mean Median SD
Ponderosa pine
Wm 0.43 0.43 0.10
WmWt 0.43 0.42 0.11
Wt 0.36 0.34 0.09
Co 0.35 0.36 0.07
Douglas-fir
Wm 0.52 0.52 0.09
WmWt 0.46 0.45 0.08
Wt 0.38 0.36 0.10
Co 0.36 0.34 0.10
Note: Data were collected for both ponderosa
pine and Douglas-fir after 2 years of experimen-
tal treatment. Results from 2x2 ANOVA indicate
that root to shoot ratios differed by treatment
type for both ponderosa pine (F= 11.98, P< 0.001)
and Douglas-fir (F= 16.17, P< 0.001). SD, standard
deviation. Treatment types: Wm, warmed only;
WmWt, warmed and watered; Wt, watered on-
ly; Co, control.
Fig. 5. Total nonconifer biomass (g·m
−2
) (i.e., grasses, forbs, and
shrubs) by treatment type based on biomass harvesting completed
at the end of year 2 (2013) of the experiment. The thick black line
inside the box indicates the median, the lines at the outer edges of
the box indicate the upper and lower quartiles, and the lines at the
end of vertical dashed lines indicate the maximum and minimum
values. The circle indicates an outlier. Results of 2×2 ANOVA
indicate that means are statistically equal. Treatment types: Wm,
warmed only; WmWt, warmed and watered; Wt, watered only; Co,
control.
Wm WmWt Wt Co
0 20406080100
Total Nonconifer Biomass (g m-2)
Treatment Type
.
1614 Can. J. For. Res. Vol. 45, 2015
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF COLORADO on 10/04/15
For personal use only.
regeneration may be restricted to areas that provide suitable mi-
croclimate conditions such as higher elevations and north-facing
slopes. It is possible that a transition to grasslands or shrublands may
occur in some areas following wildfire. Although our study focused
on low-elevation forests of the CFR, our findings are relevant to other
forested ecosystems in which increased temperatures and associated
drought may similarly inhibit post-disturbance regeneration by the
dominant tree species. We expect that many forests worldwide are
currently vulnerable to post-disturbance shifts in vegetation pat-
terns due to climate change, as supported by research in northern
Patagonia (Tercero-Bucardo et al. 2007), the Central Alps (Moser et al.
2010), the eastern US (Overpeck et al. 1990), the western US (Dodson
and Root 2013;Feddema et al. 2013;Johnstone et al. 2010b;Savage
et al. 2013), and portions of Canada (Hogg and Schwarz 1997;
Landhausser and Wein 1993). In areas where climate-mediated shifts
in vegetation following disturbance are anticipated, land managers
will need to act quickly to prioritize areas where they would like to
forestall loss of forested cover following fire vs. areas where the
persistence of alternative vegetation communities is acceptable or
desired.
Acknowledgements
This research was supported by the National Science Founda-
tion (grant nos. 1232997 and 0966472) and the Graduate Research
Fellowship Program. Boulder County Parks and Open Space
(BCPOS) also provided financial and staff support in implementing
and monitoring the experiment. We especially thank N. Stremel,
E. Duncan, and W. Foster for project planning and fieldwork as-
sistance and P. Pendergast for advice regarding statistical analy-
ses. We also thank the anonymous reviewers whose suggestions
significantly improved the manuscript.
References
Baum, C.F. 2006. An introduction to modern econometrics using Stata. Stata
Press.
Bonnet, V.H., Schoettle, A.W., and Shepperd, W.D. 2005. Postfire environmental
conditions influence the spatial pattern of regeneration for Pinus ponderosa.
Can. J. For. Res. 35(1): 37–47. doi:10.1139/x04-157.
Callaway, R.M., DeLucia, E.H., and Schlesinger, W.H. 1994. Biomass allocation of
montane and desert ponderosa pine: an analog for response to climate
change. Ecology, 75: 1474–1481. doi:10.2307/1937470.
Cleary, B.D. 1970. The effect of plant moisture stress on the physiology and
establishment of planted Douglas-fir and ponderosa pine seedlings. Oregon
State University, School of Forestry, Corvallis, Oregon.
Cooper, C.F. 1960. Changes in vegetation, structure, and growth of southwestern
pine forests since white settlement. Ecol. Monogr. 30: 129–164. doi:10.2307/
1948549.
Delucia, E.H., Maherali, H., and Carey, E.V. 2000. Climate-driven changes in
biomass allocation in pines. Glob. Chang. Biol. 6: 587–593. doi:10.1046/j.1365-
2486.2000.00338.x.
Dodson, E.K., and Root, H.T. 2013. Conifer regeneration following stand-
replacing wildfire varies along an elevation gradient in a ponderosa pine
forest, Oregon, U.S.A. For. Ecol. Manage. 302: 163–170. doi:10.1016/j.foreco.
2013.03.050.
Ebel, B.A., and Mirus, B.B. 2014. Disturbance hydrology: challenges and oppor-
tunities. Hydrol. Process. 28(19): 5140–5148. doi:10.1002/hyp.10256.
Ehle, D.S., and Baker, W.L. 2003. Disturbance and stand dynamics in ponderosa
pine forests in Rocky Mountain National Park, U.S.A. Ecol. Monogr. 73: 543–
566. doi:10.1890/03-4014.
Enright, N.J., Fontaine, J.B., Bowman, D.M.J.S., Bradstock, R.A., and Williams, R.J.
2015. Interval squeeze: altered fire regimes and demographic responses in-
teract to threaten woody species persistence as climate changes. Front. Ecol.
13: 265–272. doi:10.1890/140231.
Feddema, J.J., Mast, J.N., and Savage, M. 2013. Modeling high-severity fire,
drought and climate change impacts on ponderosa pine regeneration. Ecol.
Model. 253: 56–69. doi:10.1016/j.ecolmodel.2012.12.029.
Graham, R., Finney, M., McHugh, C., Cohen, J., Calkin, D., Stratton, R., Bradshaw, L.,
and Nikolov, N. 2012. Fourmile Canyon fire findings. USDA Forest Service,
Rocky Mountain Research Station, Fort Collins, Colorado, General Technical
Report RMRS-GTR-289.
Haire, S.L., and McGarigal, K. 2010. Effects of landscape patterns of fire severity
on regenerating ponderosa pine forests (Pinus ponderosa) in New Mexico and
Arizona, U.S.A. Landsc. Ecol. 25: 1055–1069. doi:10.1007/s10980-010-9480-3.
Hogg, E.H., and Schwarz, A.G. 1997. Regeneration of planted conifers across
climatic moisture gradients on the Canadian prairies: implications for dis-
tribution and climate change. J. Biogeogr. 24: 527–534. doi:10.1111/j.1365-2699.
1997.00138.x.
Holling, C.S. 1973. Resilience and stability of ecological systems. Annu. Rev. Ecol.
Syst. 4: 1–23. doi:10.1146/annurev.es.04.110173.000245.
Hollister, R.D., and Webber, P.J. 2000. Biotic validation of small open-top cham-
bers in a tundra ecosystem. Glob. Chang. Biol. 6: 835–842. doi:10.1046/j.1365-
2486.2000.00363.x.
Johnstone, J.F., Chapin, F.S., III, Hollingsworth, T.N., Mack, M.C.,
Romanovsky, V., and Turetsky, M. 2010a. Fire, climate change, and forest
resilience in interior Alaska. Can. J. For. Res. 40(7): 1302–1312. doi:10.1139/X10-
061.
Johnstone, J.F., McIntire, E.J.B., Pedersen, E.J., King, G., and Pisaric, M.J.F. 2010b.
A sensitive slope: estimating landscape patterns of forest resilience in a
changing climate. Ecosphere, 1: art14. doi:10.1890/ES10-00102.1.
Kaufmann, M.R., Veblen, T.T., and Romme, W.H. 2006. Historical fire regimes in
ponderosa pine forests of the Colorado Front Range, and recommendations
for ecological restoration and fuels management. Front Range Fuels Treat-
ment Partnership Roundtable, findings of the Ecological Workgroup.
Kayes, L.J., Anderson, P.D., and Puettmann, K.J. 2010. Vegetation succession
among and within structural layers following wildfire in managed forests.
J. Veg. Sci. 21: 233–247. doi:10.1111/j.1654-1103.2009.01136.x.
Keyser, T.L., Lentile, L.B., Smith, F.W., and Shepperd, W.D. 2008. Changes in
forest structure after a large, mixed-severity wildfire in ponderosa pine for-
ests of the Black Hills, South Dakota, U.S.A. For. Sci. 54: 328–338.
Landhausser, S.M., and Wein, R.W. 1993. Postfire vegetation recovery and tree
establishment at the Arctic treeline: climate-change–vegetation-response hy-
potheses. J. Ecol. 81: 665–672. doi:10.2307/2261664.
League, K., and Veblen, T. 2006. Climatic variability and episodic Pinus ponderosa
establishment along the forest–grassland ecotones of Colorado. For. Ecol.
Manage. 228: 98–107. doi:10.1016/j.foreco.2006.02.030.
Lukas, J., Barsugli, J., Doesken, N., Rangwala, I., and Wolter, K. 2014. Climate
change in Colorado: a synthesis to support water resources management and
adaptation. University of Colorado, Boulder, Colorado.
Marion, G.M., Henry, G.H.R., Freckman, D.W., Johnstone, J., Jones, G.,
Jones, M.H., Levesque, E., Molau, U., Mølgaard, P., Parsons, A.N., Svoboda, J.,
and Virginia, R.A. 1997. Open-top designs for manipulating field temperature
in high-latitude ecosystems. Glob. Chang. Biol. 3: 20–32. doi:10.1111/j.1365-
2486.1997.gcb136.x.
Mast, J.N., Veblen, T.T., and Linhart, Y.B. 1998. Disturbance and climatic influ-
ences on age structure of ponderosa pine at the pine/grassland ecotone,
Colorado Front Range. J. Biogeogr. 25: 743–755. doi:10.1046/j.1365-2699.1998.
2540743.x.
McDowell, N.G., Beerling, D.J., Breshears, D.D., Fisher, R.A., Raffa, K.F., and
Stitt, M. 2011. The interdependence of mechanisms underlying climate-
driven vegetation mortality. Trends Ecol. Evol. 26: 523–532. doi:10.1016/j.tree.
2011.06.003.
Millar, C.I., Stephenson, N.L., and Stephens, S.L. 2007. Climate change and for-
ests of the future: managing in the face of uncertainty. Ecol. Appl. 17: 2145–
2151. doi:10.1890/06-1715.1.
Montes-Helu, M.C., Kolb, T., Dore, S., Sullivan, B., Hart, S.C., Koch, G., and
Hungate, B.A. 2009. Persistent effects of fire-induced vegetation change on
energy partitioning and evapotranspiration in ponderosa pine forests. Agric.
For. Meteorol. 149: 491–500. doi:10.1016/j.agrformet.2008.09.011.
Mooney, K.A., Linhart, Y.B., and Snyder, M.A. 2011. Masting in ponderosa pine:
comparisons of pollen and seed over space and time. Oecologia, 165: 651–661.
doi:10.1007/s00442-010-1742-x.
Moser, B., Temperli, C., Schneiter, G., and Wohlgemuth, T. 2010. Potential shift
in tree species composition after interaction of fire and drought in the Cen-
tral Alps. Eur. J. For. Res. 129: 625–633. doi:10.1007/s10342-010-0363-6.
Olszyk, D.M., Johnson, M.G., Tingey, D.T., Rygiewicz, P.T., Wise, C., VanEss, E.,
Benson, A., Storm, M.J., and King, R. 2003. Whole-seedling biomass alloca-
tion, leaf area, and tissue chemistry for Douglas-fir exposed to elevated CO
2
and temperature for 4 years. Can. J. For. Res. 33(2): 269–278. doi:10.1139/x02-
186.
Overpeck, J.T., Rind, D., and Goldberg, R. 1990. Climate-induced changes in
forest disturbance and vegetation. Nature, 343: 51–53. doi:10.1038/343051a0.
Pearson, G.A. 1923. Natural reproduction of western yellow pine in the South-
west. USDA Forest Service, Washington D.C., Bulletin no. 1105.
Peet, R.K. 1981. Forest vegetation of the Colorado Front Range: composition and
dynamics. Vegetatio, 45: 3–75. doi:10.1007/BF00240202.
Reclamation. 2013. Downscaled CMIP3 and CMIP5 climate projections: release of
downscaled CMIP5 climate projections, comparison with preceding informa-
tion, and summary of user needs. U.S. Department of the Interior, Bureau of
Reclamation, Technical Service Center, Denver, Colorado. Available from
http://gdo-dcp.ucllnl.org/downscaled_cmip_projections/techmemo/
downscaled_climate.pdf [accessed 8 April 2014].
Roccaforte, J.P., Fule, P.Z., and Covington, W.W. 2010. Monitoring landscape-
scale ponderosa pine restoration treatment implementation and effective-
ness. Restor. Ecol. 18: 820–833. doi:10.1111/j.1526-100X.2008.00508.x.
Roccaforte, J.P., Fulé, P.Z., Chancellor, W.W., and Laughlin, D.C. 2012. Woody
debris and tree regeneration dynamics following severe wildfires in Arizona
ponderosa pine forests. Can. J. For. Res. 42(3): 593–604. doi:10.1139/x2012-010.
Rother et al. 1615
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF COLORADO on 10/04/15
For personal use only.
Savage, M., and Mast, J.N. 2005. How resilient are southwestern ponderosa pine
forests after crown fires? Can. J. For. Res. 35(4): 967–977. doi:10.1139/x05-028.
Savage, M., Brown, P.M., and Feddema, J. 1996. The role of climate in a pine forest
regeneration pulse in the southwestern United States. Ecoscience, 3: 310–318.
Savage, M., Mast, J.N., and Feddema, J.J. 2013. Double whammy: high-severity fire
and drought in ponderosa pine forests of the Southwest. Can. J. For. Res.
43(6): 570–583. doi:10.1139/cjfr-2012-0404.
Sevanto, S., McDowell, N.G., Dickman, L.T., Pangle, R., and Pockman, W.T. 2014.
How do trees die? A test of the hydraulic failure and carbon starvation hy-
potheses. Plant Cell Environ. 37: 153–161. doi:10.1111/pce.12141.
Shatford, J.P.A., Hibbs, D.E., and Puettmann, K.J. 2007. Conifer regeneration
after forest fire in the Klamath–Siskiyous: how much, how soon? J. For. 105:
139–146.
Sherriff, R.L., and Veblen, T.T. 2007. A spatially-explicit reconstruction of histor-
ical fire occurrence in the ponderosa pine zone of the Colorado Front Range.
Ecosystems, 10: 311–323. doi:10.1007/s10021-007-9022-2.
Spittlehouse, D.L., and Stewart, R.B. 2003. Adaptation to climate change in forest
management. BC J. Ecosyst. Manag. 4: 1–11.
StataCorp. 2015. Stata statistical software: release 14. StataCorp LP, College Sta-
tion, Texas.
Tercero-Bucardo, N., Kitzberger, T., Veblen, T.T., and Raffaele, E. 2007. A field
experiment on climatic and herbivore impacts on post-fire tree regeneration
in north–western Patagonia. J. Ecol. 95: 771–779. doi:10.1111/j.1365-2745.2007.
01249.x.
Turnball, M.H., Murthy, R., and Griffin, K.L. 2002. The relative impacts of day-
time and night-time warming on photosynthetic capacity in Populus deltoides.
Plant Cell Environ. 25: 1729–1737. doi:10.1046/j.1365-3040.2002.00947.x.
Turner, M.G. 2010. Disturbance and landscape dynamics in a changing world.
Ecology, 91: 2833–2849. doi:10.1890/10-0097.1.
Turner, N.C. 1997. Further progress in crop water relations. Adv. Agron. 58:
293–338. doi:10.1016/S0065-2113(08)60258-8.
Ulery, A.L., and Graham, R.C. 1993. Forest fire effects on soil color and texture. Soil
Sci. Soc. Am. J. 57: 135–140. doi:10.2136/sssaj1993.03615995005700010026x.
van Mantgem, P.J., Stephenson, N.L., Byrne, J.C., Daniels, L.D., Franklin, J.F.,
Fule, P.Z., Harmon, M.E., Larson, A.J., Smith, J.M., Taylor, A.H., and
Veblen, T.T. 2009. Widespread increase of tree mortality rates in the western
United States. Science, 323: 521–524. doi:10.1126/science.1165000.
Veblen, T.T., and Lorenz, D.C. 1986. Anthropogenic disturbance and recovery
patterns in montane forests, Colorado Front Range. Prog. Phys. Geogr. 7:
1–24.
Williams, A.P., Allen, C.D., Macalady, A.K., Griffin, D., Woodhouse, C.A.,
Meko, D.M., Swetnam, T.W., Rauscher, S.A., Seager, R., Grissino-Mayer, H.D.,
Dean, J.S., Cook, E.R., Gangodagamage, C., Cai, M., and McDowell, N.G. 2013.
Temperature as a potent driver of regional forest drought stress and tree
mortality. Nat. Clim. Change, 3: 292–297. doi:10.1038/nclimate1693.
1616 Can. J. For. Res. Vol. 45, 2015
Published by NRC Research Press
Can. J. For. Res. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF COLORADO on 10/04/15
For personal use only.
