Directory UMM :Data Elmu:jurnal:T:Tree Physiology:Vol14.1994:
Tree Physiology 14, 1339--1349
© 1994 Heron Publishing----Victoria, Canada
Effect of elevated temperatures on carbon dioxide exchange in
Picea rubens
DAVID R. VANN,1 ARTHUR H. JOHNSON2 and BRENDA B. CASPER1
1
Department of Biology, University of Pennsylvania, PA 19104, USA
2
Department of Geology, University of Pennsylvania, PA 19104, USA
Received August 25, 1993
Summary
We examined some of the physiological reasons that may underlie past and expected future migrations
of red spruce (Picea rubens Sarg.) by evaluating the effects of high temperatures on photosynthesis and
respiration of trees growing on Whiteface Mountain, NY. At temperatures of 35--40 °C, the trees
exhibited a zero or negative carbon balance. Higher temperatures resulted in cellular disorganization and
death. Temperatures around 30 °C resulted in reduced CO2 uptake, a condition that could decrease future
reproductive output and competitive stature. We conclude that thermal intolerance explains, at least in
part, the absence of red spruce at low elevations and latitudes where temperatures of ≥ 30 °C occur. We
suggest that the thermosensitivity of this species is important with respect to global climate trends and
migration patterns.
Keywords: global warming, migration patterns, photosynthesis, red spruce, respiration, thermotolerance.
Introduction
Heat stress in trees has been studied in relatively thermotolerant species that do not
experience injury or death until temperatures are greater than 45 °C (see Levitt 1980,
Kozlowski et al. 1991). Cold-temperate and holarctic tree species may be less
thermotolerant, because photosynthesis in these plants declines at temperatures
around 30 °C (Larcher 1969). These species, principally Picea and Abies, followed
the northward retreat of the Holocene ice sheets and currently inhabit the cooler
arctic, boreal and upper montane regions (Davis et al. 1980). Competition with faster
growing angiosperms has been cited as a reason for the decline of conifers in
temperate regions (Davis and Botkin 1985); however, these conifers evolved under
cool conditions and so may not be adapted to the warmer conditions that prevail in
angiosperm dominated areas. If this is true, boreal species would be expected to be
less thermotolerant than other species.
Red spruce (Picea rubens Sarg.) is a model species in which to test this hypothesis.
It has a restricted range, being indigenous to the Appalachian mountains (Blum
1990). Its lower elevation limits vary inversely with latitude, essentially coinciding
with the 27.7 °C isotherm for average July daily maximum temperature (Environmental Data Services 1968). Red spruce migrated from Pleistocene refuges in more
southerly coastal areas, with its current distribution becoming established by the
early Holocene (Whitehead 1973, Watts 1979, Ibe 1985). There is evidence that its
1340
VANN, JOHNSON AND CASPER
range was further altered during the mid-Holocene warming events (Whittaker 1956,
Mark 1958). These historic events suggest a sensitivity to warm temperatures.
As a part of a three-year investigation of the effects of airborne chemicals on red
spruce, branches were fitted with environmental chambers (Vann and Johnson 1988,
Vann et al. 1992). During 1987 and 1988, technical problems resulted in elevated
temperatures in some chambers for short periods. In some instances, the air temperatures rose as much as 10 °C above ambient, to a maximum of 42 °C. The highest
temperatures occurred during the noon hour of greatest solar exposure. Although air
temperatures in the chambers remained below 35 °C for most of the time, some
needles exhibited chlorosis and reddening. This visible injury occurred either as a
result of a short exposure to temperatures seldom occurring in nature at this site
(> 37 °C), or as a result of prolonged exposure to temperatures between 32 and
37 °C. In both years, the greatest visible injury occurred on chambered branches of
trees that were declining (i.e., had extensive canopy needle loss), with less visible
damage on chambered branches of healthy trees.
Because the symptoms appeared to be due to chlorophyll loss, we examined the
pigment concentration and anatomy of the damaged tissues. Because visible mortality of tissues occurred at temperatures lower than those reported for other species
(i.e., below 40 °C), we hypothesized that trees in the field would experience a net
reduction in photosynthesis at temperatures common at lower elevations. To test this
hypothesis, we measured gas exchange rates of shoots on trees at both ambient and
elevated temperatures. Light and dark CO2 exchange rates were used to calculate
photosynthetic and respiration rates.
Materials and methods
The chamber experiments performed in 1987 and 1988 used branch-sized environmental chambers that displaced approximately 85 l. The chambers were continuously
evacuated with fans displacing about 500 l min −1, and the air in each chamber was
mixed with two small fans. Details of the construction, operation and experimental
design are presented elsewhere (Vann and Johnson 1988).
Chlorophyll determinations
A set of six sample twigs, three visibly chlorotic and three with no visible signs of
injury, were cut from a chambered branch of the tree showing injury on July 16, 1987,
four days after overheating. Excised shoots were frozen in liquid nitrogen, transported in dry ice and ground in liquid nitrogen. Three 50-mg replicates were then
extracted twice in 5 ml of 100% acetone and analyzed spectrophotometrically
(Beckman DU-50) at 478, 645 and 662 nm. Chlorophyll concentrations were calculated based on the formulas of Wellburn and Lichtenthaler (1984). Because the water
content of injured tissues was at least 15% less than that of uninjured tissues,
calculations were based on dry weight. Comparisons of mean chlorophyll concentrations were based on the t-test.
HEAT AND CARBON DIOXIDE EXCHANGE IN RED SPRUCE
1341
Electron microscopy
Sample needles were removed from a single shoot having both green and chlorotic
needles. Each needle was sliced under fixative into 2--4 mm sections. The fixative
contained 2% formaldehyde and 2.5% glutaraldehyde in PIPES buffer (Salerma and
Brandaõ 1973), with 0.5% caffeine added to precipitate vacuolar tannins in situ
(Müller and Rodehurst 1977). The slices were placed in small vials of fixative,
lowered from the tree and vacuum infiltrated. Vials were transported to the laboratory
in ice. After fixing for 12 h, samples were transferred through two changes of 0.1 M
PIPES buffer (45 min each), soaked in 2% osmium tetroxide at pH 6.8 for 2 h, and
dehydrated through an ethanol series to propylene oxide (PO) according to standard
procedures (Hayat 1981). Embedding was performed by transferring tissues through
three changes of decreasing ratios of PO/Spurr’s resin to pure Spurr’s resin (Spurr
1969). Specimens in pure resin were held for 12 h, then transferred to fresh resin and
polymerized at 70 °C. Blocks were sectioned at 90 µm on an AO Reichert microtome
and examined with a Zeiss 10 electron microscope.
Gas exchange
To assess the extent to which high temperatures affect the tree before the onset of
visible injury, we examined CO2 exchange rates at ambient and elevated temperatures. Two experiments were conducted, both at Whiteface Mountain, NY. The initial
experiment was conducted in September 1989 on two mature, canopy dominant trees
at the 1170 m site described elsewhere (Vann and Johnson 1988, Vann et al. 1992).
Two branches were selected in the western facing sector of each tree. The trees were
approximately 27 and 15 m tall. The larger tree had lost most of its canopy, and was
in a state of severe decline. A branch was selected from the eighth whorl from the top
of the tree. The smaller tree was visibly healthy, and a branch from the fifth whorl
was used. For the second experiment in early August 1990, a group of four trees, all
between 1.5 and 2 m in height and 3--4 cm in diameter at the base, were selected at
the 900 m site. Current-year shoots, comprising the twig and 50--70 needles, were
selected at random from exposed accessible branches on the western side of the trees.
Gas exchange rates of these shoots were determined with a portable infrared gas
analyzer (ADC model LCA-2). Air was supplied and controlled by means of an ADC
model ASUM2 pump. This equipment was fitted to one of two custom-made
cuvettes, depending on whether current-year or 1-year-old foliage was being measured. In the 1989 experiment, only current-year foliage was measured. The cuvettes,
which were designed to enclose single shoots, had small volumes (< 100 cm3). On
completion of all tests, the shoots were cut and weighed. Needle dry weight was used
to standardize the measured gas exchange rates.
Temperatures within the cuvettes were controlled by decreasing the flow rate in the
chamber. Initial measurements were made at a flow rate of 500 l min −1, resulting in
a chamber air volume turnover rate of about six times per minute. This rate was
necessary to maintain the chamber temperature near ambient, and was probably a
result of the small chamber size (< 1 cm larger in diameter than the shoot). The
1342
VANN, JOHNSON AND CASPER
temperature was increased by reducing the flow to 100 l min −1. Slower flows resulted
in a temperature increase that was too rapid to allow determination of a stable
photosynthetic rate, and faster flows took too long to reach the desired temperatures.
Use of flow rate to adjust temperature was chosen for two reasons. First, it was more
convenient for field determinations, because it required no additional equipment.
Second, increased temperature can result in an increase in the water vapor pressure
gradient (Kramer 1983). Some of the effect of heat on photosynthesis may be a result
of desiccation caused by this increase in the vapor deficit (Larcher 1983). Reducing
the air flow permitted the chamber air to become saturated with water vapor from
transpiration, thus minimizing effects arising from desiccation. Although altering the
flow rates may be expected to alter the boundary layer size, calculations (based on
equations in Nobel 1991) suggest that the actual change was less than 10%. Temperature was determined by placing a microthermocouple against the underside of a
single needle in the upper portion of the enclosed shoot; all temperatures reported are
thus needle surface temperatures.
Once the chamber was installed, the twig was given about 10 min for gas exchange
to stabilize. Two readings were then taken 5 min apart for the determination of
photosynthetic rate at ambient temperature. The flow was then decreased and the
chamber was covered with a sleeve of black plastic and shaded. Five min later,
respiration rate was determined, and the shading and sleeve were removed. The flow
rate was held low until the chamber temperature reached about 40 °C. One measurement was taken after about 15--30 min had elapsed, when the temperature was in the
mid-30 °C range, and a second measurement was taken when the temperature had
reached about 40 °C. To obtain a duplicate measurement, the branch was held at
about 40 °C for 15 min by intermittently shading the chamber. The flow rate was then
increased to 1 l min −1 for 30 s to flush the chamber, then returned to the low flow rate
(needed to improve the resolution of the respiration rate measurement). The chamber
was then covered with a black plastic sleeve, and the temperature permitted to rise to
about 40 °C. As it was not possible to hold the chamber at a constant temperature for
more than a few minutes at a time, two measurements were taken 2 min apart to
determine the respiration rate at the elevated temperature.
In 1989, measurements were made on September 9; in 1990, current-year foliage
was measured on August 3 and 4, and 1-year-old foliage was tested 1 week later.
Because it was desirable to maintain consistent light conditions during the experiment, measurements were conducted only when the site was in direct sun, a period
of a few hours in the early afternoon. This kept the shoots under study in saturating
light conditions and in appropriate conditions for solar heating. Because it took about
20 min to install and check the chamber for leaks, and over an hour to run the test, it
was possible to analyze only two trees per day. Measurements on 1-year-old foliage
were obtained for only one branch per tree. On subsequent days, four of the
current-year twigs that had been tested the previous week were remeasured to
determine recovery.
During the 1989 experiment, the mountain had a light cloud cover, reducing
HEAT AND CARBON DIOXIDE EXCHANGE IN RED SPRUCE
1343
ambient irradiance to about 230 µmol m −2 s −1. Although this is well below the light
saturation value, it is adequate for measurable carbon gain in this species (Manley
and Ledig 1979). The 1990 experiment was conducted on cloudless days with an
irradiance of about 1800 µmol m −2 s −1, well above the light saturation point (Manley
and Ledig 1979, McLaughlin et al. 1990).
Results
In 1987, there was a loss of chlorophyll and carotene in the visibly injured cells of
red spruce needles (Figure 1). The change was most pronounced in the 1-year-old
(1986) needle tissue. Visibly injured 1-year-old needles lost 45% of their chlorophyll
compared with uninjured needles from nearby shoots (T ≈ 13.1, P < 0.002). Carotenoids also decreased significantly, but to a lesser degree than chlorophyll (37.8%, T
≈ 10, P < 0.006). The current-year (1987) needles also lost significant amounts of
chlorophyll (34.6%, T ≈ 11.9, P < 0.003) and carotenoids (26.5%, T ≈ 3.46, P < 0.03).
The loss of chlorophyll in chlorotic tissues resulted from the breakdown of
chloroplast membranes (Figures 2A and 2B). Figure 2D shows a chloroplast with a
normal membrane system taken from a green needle in the same chamber. In the
damaged needles, the thylakoids had disintegrated, leaving only lipid droplets (od)
(cf. Levitt 1980). In addition, the chloroplasts became more spherical in shape.
Injured cells also exhibited an altered precipitation of vacuolar contents (Figure 2C).
Although the relationship between vacuolar chemistry and the visible structural
differences is not understood, the visible changes were consistent in all cells examined, and may reflect disruptions of the cellular osmotic environment. In some cells,
cellular structures had precipitated completely, possibly as a result of tonoplast
rupture. Needles that were visibly damaged became increasingly necrotic during the
following two weeks, and eventually abscised.
Figure 1. Effect of elevated air temperature on pigment concentrations of current and 1-year-old needles
of red spruce sampled in 1987. U = uninjured needles, I = visibly injured, j = chlorophyll, and h =
carotenoids. Bars = 95% c.i.
1344
VANN, JOHNSON AND CASPER
Figure 2. Electron micrographs of heat-damaged (A--C) and undamaged needle cells (D--E). (A) cell,
(B) chloroplast detail, (C) vacuole detail, (D) chloroplast detail, and (E) cell. Legend: n = nucleus, m =
mitochondrion, cw = cell wall, sg = starch grain, og = osmiophilic globule, od = oil droplet, nvt = normal
vacuolar tannin, and pvt = precipitated vacuolar tannin.
There was a rapid decrease in photosynthetic rate at temperatures above 30 °C
(Figure 3). At temperatures around 40 °C, gross photosynthetic rates declined below
the CO2 compensation point. Photosynthetic rates in 1-year-old needles were about
30% lower than in current-year needles (Figure 3B). Figure 3A shows data for
current-year needles for both 1989 and 1990. The initial values below 20 °C were
from the trees measured in 1989 under overcast skies. The regression between the
temperature response (Tmp) and photosynthetic rate (Ps) in current-year needles had
a steeper slope in 1990 than in 1989. The equation for the 1989 current-year needles
was Ps = 7.25 − 0.225Tmp (R2 = 0.753, P < 0.01), and for the 1990 current-year
needles was Ps = 12.3 − 0.331Tmp (R2 = 0.952, P < 0.01). The response of 1-year-old
tissue was similar: Ps = 7.79 − 0.234Tmp (R2 = 0.925, P < 0.01). The low R2 value in
the 1989 current-year foliage presumably reflects differences between the trees; one
was declining, the other visibly healthy. The high R2 values for the trees measured in
1990 may be a consequence of selecting trees that were similar. All tissues declined
toward a common value of zero, reducing the variance at high temperatures.
Respiration approximately doubled over the temperature range of 10 to 15 °C. This
HEAT AND CARBON DIOXIDE EXCHANGE IN RED SPRUCE
1345
Figure 3. Effect of temperature on CO2 exchange in (A) current-year needles and (B) 1-year-old needles
of red spruce. Apparent photosynthesis at ambient temperatures = d, at 10 °C above ambient = e, and
at 15 °C above ambient = j. Respiration at ambient temperature = n, and at 15 °C above ambient = ,.
Shaded symbols denote data collected in 1989; other symbols denote data collected in 1990. Temperatures are needle surface temperatures.
measurement may be low, because respiration in this species is low, approaching the
sensitivity of the instrument used. To increase the ∆CO2 value, the respiration
measurements were determined at very low flow rates. This resulted in high chamber
humidities accompanied by some condensation in the exhaust tube, and may have
altered CO2 concentrations through stomatal closure or through dissolution into the
water droplets. Respiration rates increased to similarly high values in both years,
although the temperatures were not comparable. The regression between temperature
response (Tmp) and respiration rate (Rs) was performed individually on each year’s
data. The equations obtained for the current-year needles were Rs = 0.347 − 0.089Tmp
(R2 = 0.955, P < 0.01) and Rs = 1.30 − 0.089Tmp (R2 = 0.937, P < 0.01) for 1989 and
1990, respectively. The 1-year-old needles exhibited a more gradual response; the
equation was Rs = − 0.434 − 0.036Tmp (R2 = 0.68, P < 0.01). Although the rate of
response in 1-year-old needles was unexpectedly low, given that these tissues were
the first to become visibly damaged in response to overheating, the respiration rates
at the higher temperature were significantly different from those at the ambient
temperature (T = 3.84, P < 0.01). The measurements with 1-year-old needles covered
a smaller temperature range than the current-year tests, partly because of shading
from adjacent twigs. Respiration of the 1-year-old needles may have increased
nonlinearly at higher temperatures. Because the cuvettes were small and could only
accommodate a single shoot (typically 3--4.5 cm in length), and flow rate was low,
the boundary layer may have been relatively large, thereby reducing the accuracy of
1346
VANN, JOHNSON AND CASPER
the measurement.
The second set of measurements taken in 1990 on current-year foliage were 92 ±
7% of the values determined the week before (data not shown). These values were
taken to indicate complete recovery of the photosynthetic capacity from overheating;
some of the difference in rates was likely due to different weather conditions on the
day of measurement.
Discussion
Manipulation of chamber temperatures did not precisely mimic conditions in nature,
because the humidity in the chamber remained near saturation. In nature, the plant
would lose water vapor freely. This permits cooling, but under some conditions could
result in water stress, which would exacerbate the effects of heat on photosynthesis
(Levitt 1980, Kramer 1983, Kozlowski et al. 1991). Although transpirational cooling
may reduce heating, conifer needles typically track ambient temperatures when there
is wind (Tibbals et al. 1964), and red spruce needles in full sun with little wind may
be 3--4 °C warmer than the air (D. Vann, unpublished data). In addition, transpiration
rates are low in conifers, resulting in decreases in needle temperature due to evaporative cooling of only 1--3 °C (Tibbals et al. 1964). Therefore, it is not clear whether
there is sufficient transpirational cooling in this species to protect it from injury at
high temperatures. Trees at low elevations may often be exposed to injurious needle
temperatures, particularly in the southern Appalachians where air temperatures often
reach 35--37 °C (Environmental Data Services 1968). In this experiment, injuries
were presumed to have occurred primarily or wholly as a result of direct heating,
because the high humidity must have reduced transpirational cooling.
Visible chlorosis, indicating severe cellular injury, occurred between 32 and 40 °C.
This is about 20 °C lower than the range of 50--55 °C reported for heat shock injury
in temperate plants, and about 10 °C below values causing damage after long-term
exposure (see compilations in Levitt 1980, Kozlowski et al. 1991). The needle tissues
damaged were generally from trees that were declining, and may have been sensitized to additional stress.
Temperatures can frequently reach values that reduce the net carbon gain of the
plant, but do not cause tissue death. Above 30--35 °C, red spruce trees fix little, if
any, carbon. The temperature compensation point values found for red spruce in this
study are similar to those reported for other cold temperate conifers (Larcher 1969,
Fryer and Ledig 1972). It should be noted that the short duration of the heat treatment
in this experiment likely underestimated the actual effect of high temperatures on
carbon gain, because exposure times in nature might last several hours per day. The
decrease in CO2 uptake in response to elevated temperatures suggests that a reduction
in growth will occur if temperatures are elevated for long periods (cf. Levitt 1980).
This prediction is consistent with evidence from dendrochronological studies in red
spruce indicating that unusually warm August temperatures are followed by smaller
rings in the following year (Cook and Zedaker 1992).
HEAT AND CARBON DIOXIDE EXCHANGE IN RED SPRUCE
1347
Although linear regressions were used to evaluate the decline in Ps, this does not
demonstrate that the decline was linear with increasing temperature. Data were not
taken at a sufficient range of temperatures to establish the shape of the curve;
however, published data suggest that such curves are parabolic (Larcher 1969). The
higher photosynthetic rates at 25 °C may indicate an optimum, but may also be
related to the age (youth, vigor) of the trees measured in 1990 and cannot be regarded
as a continuation of those measured at 15 °C. For this reason, a parabolic curve was
not fitted to the whole data set, although it does appear to correspond with maxima
suggested by the data of Manley and Ledig (1979).
Current-year needle respiration rates at elevated temperatures were similar in both
years, although the maximum temperatures attained differed by almost 10 °C. This
may be due to an inherent maximum respiration rate due to enzymatic limitations, or
may represent a decrease at the higher (near 40 °C) temperature as a consequence of
the breakdown of cellular components. Alternately, it may be related to the different
trees used in the two experiments; compared with the trees measured in 1990, the
1989 trees were 75--100 years older, 270 m higher in elevation and, in one case,
declining. All of these factors may influence respiration rates. The elevational effect
may explain the significantly (T = 2.78, P < 0.02) higher respiration rates in the
high-elevation trees, which were measured at a lower ambient temperature than the
low-elevation trees. However, based on an extrapolation of the regression for temperature increase, and assuming the high-elevation trees had been measured at an
ambient temperature of 25 °C, respiration rates would still have been higher in the
high-elevation trees than in the low-elevation trees, suggesting that that high-elevation trees may be more sensitive to high temperatures than low-elevation trees.
We conclude that the effect of elevated temperature on red spruce is an important
consideration in explaining its geographical range, and suggest that red spruce’s
historic migrations may have been driven, at least in part, by the reduction in growth
and reproduction accompanying an unfavorable carbon balance at warm temperatures.
Modeling efforts based on past distributions reconstructed from the palynological
record and current ecology show that small increases in temperature may change the
distribution of red spruce dramatically (Davis and Botkin 1985). Based on studies
with a model designed to extrapolate the consequences of global climate warming,
Overpeck et al. (1991) concluded that during the next 2--5 centuries, red spruce
would be eliminated south of about latitude 45° N. Ecotypic variants for optimal
photosynthetic temperatures have been described in balsam fir (Fryer and Ledig
1972). Similar variants may exist in red spruce, because photosynthetic optima have
been shown to respond to prior conditions (Manley and Ledig 1979). Thus, the
species may adapt to changing climates; however, each ecotype has some genetic
limit to its adaptability. Therefore, the species must rely on reproduction to migrate
out of unfavorable habitats. Based on data in Randall (1974) and Blum (1990), the
migration rate of red spruce is estimated to be about 100 m year −1. Over the
period described by Overpeck et al. (1991), red spruce may move 20--50 km. Even
1348
VANN, JOHNSON AND CASPER
assuming that this estimate is conservative, only the northernmost populations of red
spruce would survive. Thus, susceptibility to high temperatures combined with a low
migration rate may make red spruce an early victim of global warming.
Acknowledgments
We thank Scott D. Russell of the University of Oklahoma and his staff for the electron micrographs. We
also thank Julian L. Hadley for helpful comments and review of the manuscript. This research was funded
in part by a grant from the Andrew Mellon Foundation.
References
Blum, B.B. 1990. Picea rubens Sarg. Red spruce. In Silvics of North America: 1. Conifers. Agric.
Handbook No. 674. Eds. R.M. Burns and B.H. Honkala. USDA Forest Service, Washington, D.C.,
pp 250--259.
Cook, E.R. and S.M. Zedaker. 1992. The dendroecology of red spruce decline. In Ecology and Decline
of Red Spruce in the Eastern United States. Eds. C. Eager and M.B. Adams. Springer-Verlag, New
York, pp 192--231.
Davis, M.B. and D.B. Botkin. 1985. Sensitivity of cool temperate forests and their fossil pollen to rapid
climatic change. Quat. Res. 23:327--340.
Davis, M.B., R.W. Spear and L.C.K. Shane. 1980. Holocene climate of New England. Quat. Res.
14:240--250.
Environmental Data Services. 1968. Climatic atlas of the United States. US Dept. Commerce, Washington, D.C., 80 p.
Fryer, J.H. and F.T. Ledig. 1972. Microevolution of the photosynthetic temperature optimum in relation
to the elevational complex gradient. Can. J. Bot. 50:1231--1235.
Hayat, M.A. 1981. Principles and techniques of electron microscopy. University Park Press, Baltimore,
522 p.
Ibe, R.A. 1985. Postglacial montane vegetational history around Balsam Lake, Catskill Mountains, New
York. Bull. Torr. Bot. 112:176--186.
Kozlowski, T.T., P.J. Kramer and S.G. Pallardy. 1991. The physiological ecology of woody plants. In
Physiological Ecology Series. Ed. H.A. Mooney. Academic Press, San Diego, CA, 657 p.
Kramer, P.J. 1983. Water relations of plants. Academic Press, Orlando, FL, 489 p.
Larcher, W. 1969. The effect of environmental and physiological variables on the carbon dioxide
exchange of trees. Photosynthetica 3:167--198.
Larcher, W. 1983. Physiological plant ecology. Springer-Verlag, Berlin, 303 p.
Levitt, J. 1980. Responses of plants to environmental stresses: chilling, freezing, and high temperature
stresses. In Physiological Ecology Series. Ed. J. Levitt. Academic Press, Orlando, FL, 497 p.
Manley, S.A.M. and F.T. Ledig. 1979. Photosynthesis in black and red spruce and their hybrid derivatives:
ecological isolation and hybrid adaptive inferiority. Can. J. Bot. 57:305--314.
Mark, A.F. 1958. The ecology of the southern Appalachian grass balds. Ecol. Monogr. 28:293--336.
McLaughlin, S.B., C.P. Andersen, N.T. Edwards, W.K. Roy and P.A. Layton. 1990. Seasonal patterns of
photosynthesis and respiration of red spruce saplings from two elevations in declining southern
Appalachian stands. Can. J. For. Res. 20:485--495.
Müller, W.C. and E. Rodehurst. 1977. The effect of some alkaloids on the ultrastructure of phenolic
containing cells in the endodermis of cotton roots. In Proc. 35th Annual EMSA Meetings, Baton
Rouge, LA, 544 p.
Nobel, P.S. 1991. Physicochemical and environmental plant physiology. Academic Press, San Diego,
CA, pp 396--398.
Overpeck, J.T., P.J. Bartlein and T. Webb, III. 1991. Potential magnitude of future vegetation change in
eastern North America: comparisons with the past. Science 254:692--695.
Randall, A.G. 1974. Seed dispersal into two spruce--fir clear-cuts in eastern Maine. Univ. Maine Life Sci.
Agric. Exp. Stn. Res. Life Sci. 21:15.
HEAT AND CARBON DIOXIDE EXCHANGE IN RED SPRUCE
1349
Salerma, J.T. and M.R. Brandaõ. 1973. The use of PIPES buffer in the fixation of plant cells for electron
microscopy. J. Submicroscop. Cytol. 5:79--83.
Spurr, A.R. 1969. A low-viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastruct. Res. 26:31--42.
Tibbals, E.C., E.K. Carr, D.M. Gates and F. Kreith. 1964. Radiation and convection in conifers. Am. J.
Bot. 51:529--538.
Vann, D.R. and A.H. Johnson. 1988. Design and testing of a field branch enclosure for the exclusion of
atmospheric components. In Proc. on Effects of Atmospheric Pollutants on the Spruce--Fir Forests of
the Eastern United States and the Federal Republic of Germany. USDA Forest Service, Northeast
Forest Exp. Stn., Broomall (Radnor), PA, pp 431--439.
Vann, D.R., G.R. Strimbeck and A.H. Johnson. 1992. Effects of ambient levels of airborne chemicals on
freezing resistance of red spruce foliage. For. Ecol. Manage. 51:69--79.
Watts, W.A. 1979. Late Quarternary vegetation of central Appalachia and the New Jersey coastal plain.
Ecol. Monogr. 49:427--469.
Wellburn, A.H. and H. Lichtenthaler. 1984. Formulae and program to determine total carotenoids and
chlorophylls a and b of leaf extracts in different solvents. In Advances in Photosynthesis Research,
Vol. 2. Ed. C. Sybesma. Martinus Nijhoff/Dr. Junk Publishers, The Hague, Netherlands, pp 9--12.
Whitehead, D.R. 1973. Late-Wisconsin vegetational changes in unglaciated eastern North America.
Quat. Res. 3:621--631.
Whittaker, R.H. 1956. Vegetation of the Great Smoky Mountains. Ecol. Monogr. 26:1--80.
© 1994 Heron Publishing----Victoria, Canada
Effect of elevated temperatures on carbon dioxide exchange in
Picea rubens
DAVID R. VANN,1 ARTHUR H. JOHNSON2 and BRENDA B. CASPER1
1
Department of Biology, University of Pennsylvania, PA 19104, USA
2
Department of Geology, University of Pennsylvania, PA 19104, USA
Received August 25, 1993
Summary
We examined some of the physiological reasons that may underlie past and expected future migrations
of red spruce (Picea rubens Sarg.) by evaluating the effects of high temperatures on photosynthesis and
respiration of trees growing on Whiteface Mountain, NY. At temperatures of 35--40 °C, the trees
exhibited a zero or negative carbon balance. Higher temperatures resulted in cellular disorganization and
death. Temperatures around 30 °C resulted in reduced CO2 uptake, a condition that could decrease future
reproductive output and competitive stature. We conclude that thermal intolerance explains, at least in
part, the absence of red spruce at low elevations and latitudes where temperatures of ≥ 30 °C occur. We
suggest that the thermosensitivity of this species is important with respect to global climate trends and
migration patterns.
Keywords: global warming, migration patterns, photosynthesis, red spruce, respiration, thermotolerance.
Introduction
Heat stress in trees has been studied in relatively thermotolerant species that do not
experience injury or death until temperatures are greater than 45 °C (see Levitt 1980,
Kozlowski et al. 1991). Cold-temperate and holarctic tree species may be less
thermotolerant, because photosynthesis in these plants declines at temperatures
around 30 °C (Larcher 1969). These species, principally Picea and Abies, followed
the northward retreat of the Holocene ice sheets and currently inhabit the cooler
arctic, boreal and upper montane regions (Davis et al. 1980). Competition with faster
growing angiosperms has been cited as a reason for the decline of conifers in
temperate regions (Davis and Botkin 1985); however, these conifers evolved under
cool conditions and so may not be adapted to the warmer conditions that prevail in
angiosperm dominated areas. If this is true, boreal species would be expected to be
less thermotolerant than other species.
Red spruce (Picea rubens Sarg.) is a model species in which to test this hypothesis.
It has a restricted range, being indigenous to the Appalachian mountains (Blum
1990). Its lower elevation limits vary inversely with latitude, essentially coinciding
with the 27.7 °C isotherm for average July daily maximum temperature (Environmental Data Services 1968). Red spruce migrated from Pleistocene refuges in more
southerly coastal areas, with its current distribution becoming established by the
early Holocene (Whitehead 1973, Watts 1979, Ibe 1985). There is evidence that its
1340
VANN, JOHNSON AND CASPER
range was further altered during the mid-Holocene warming events (Whittaker 1956,
Mark 1958). These historic events suggest a sensitivity to warm temperatures.
As a part of a three-year investigation of the effects of airborne chemicals on red
spruce, branches were fitted with environmental chambers (Vann and Johnson 1988,
Vann et al. 1992). During 1987 and 1988, technical problems resulted in elevated
temperatures in some chambers for short periods. In some instances, the air temperatures rose as much as 10 °C above ambient, to a maximum of 42 °C. The highest
temperatures occurred during the noon hour of greatest solar exposure. Although air
temperatures in the chambers remained below 35 °C for most of the time, some
needles exhibited chlorosis and reddening. This visible injury occurred either as a
result of a short exposure to temperatures seldom occurring in nature at this site
(> 37 °C), or as a result of prolonged exposure to temperatures between 32 and
37 °C. In both years, the greatest visible injury occurred on chambered branches of
trees that were declining (i.e., had extensive canopy needle loss), with less visible
damage on chambered branches of healthy trees.
Because the symptoms appeared to be due to chlorophyll loss, we examined the
pigment concentration and anatomy of the damaged tissues. Because visible mortality of tissues occurred at temperatures lower than those reported for other species
(i.e., below 40 °C), we hypothesized that trees in the field would experience a net
reduction in photosynthesis at temperatures common at lower elevations. To test this
hypothesis, we measured gas exchange rates of shoots on trees at both ambient and
elevated temperatures. Light and dark CO2 exchange rates were used to calculate
photosynthetic and respiration rates.
Materials and methods
The chamber experiments performed in 1987 and 1988 used branch-sized environmental chambers that displaced approximately 85 l. The chambers were continuously
evacuated with fans displacing about 500 l min −1, and the air in each chamber was
mixed with two small fans. Details of the construction, operation and experimental
design are presented elsewhere (Vann and Johnson 1988).
Chlorophyll determinations
A set of six sample twigs, three visibly chlorotic and three with no visible signs of
injury, were cut from a chambered branch of the tree showing injury on July 16, 1987,
four days after overheating. Excised shoots were frozen in liquid nitrogen, transported in dry ice and ground in liquid nitrogen. Three 50-mg replicates were then
extracted twice in 5 ml of 100% acetone and analyzed spectrophotometrically
(Beckman DU-50) at 478, 645 and 662 nm. Chlorophyll concentrations were calculated based on the formulas of Wellburn and Lichtenthaler (1984). Because the water
content of injured tissues was at least 15% less than that of uninjured tissues,
calculations were based on dry weight. Comparisons of mean chlorophyll concentrations were based on the t-test.
HEAT AND CARBON DIOXIDE EXCHANGE IN RED SPRUCE
1341
Electron microscopy
Sample needles were removed from a single shoot having both green and chlorotic
needles. Each needle was sliced under fixative into 2--4 mm sections. The fixative
contained 2% formaldehyde and 2.5% glutaraldehyde in PIPES buffer (Salerma and
Brandaõ 1973), with 0.5% caffeine added to precipitate vacuolar tannins in situ
(Müller and Rodehurst 1977). The slices were placed in small vials of fixative,
lowered from the tree and vacuum infiltrated. Vials were transported to the laboratory
in ice. After fixing for 12 h, samples were transferred through two changes of 0.1 M
PIPES buffer (45 min each), soaked in 2% osmium tetroxide at pH 6.8 for 2 h, and
dehydrated through an ethanol series to propylene oxide (PO) according to standard
procedures (Hayat 1981). Embedding was performed by transferring tissues through
three changes of decreasing ratios of PO/Spurr’s resin to pure Spurr’s resin (Spurr
1969). Specimens in pure resin were held for 12 h, then transferred to fresh resin and
polymerized at 70 °C. Blocks were sectioned at 90 µm on an AO Reichert microtome
and examined with a Zeiss 10 electron microscope.
Gas exchange
To assess the extent to which high temperatures affect the tree before the onset of
visible injury, we examined CO2 exchange rates at ambient and elevated temperatures. Two experiments were conducted, both at Whiteface Mountain, NY. The initial
experiment was conducted in September 1989 on two mature, canopy dominant trees
at the 1170 m site described elsewhere (Vann and Johnson 1988, Vann et al. 1992).
Two branches were selected in the western facing sector of each tree. The trees were
approximately 27 and 15 m tall. The larger tree had lost most of its canopy, and was
in a state of severe decline. A branch was selected from the eighth whorl from the top
of the tree. The smaller tree was visibly healthy, and a branch from the fifth whorl
was used. For the second experiment in early August 1990, a group of four trees, all
between 1.5 and 2 m in height and 3--4 cm in diameter at the base, were selected at
the 900 m site. Current-year shoots, comprising the twig and 50--70 needles, were
selected at random from exposed accessible branches on the western side of the trees.
Gas exchange rates of these shoots were determined with a portable infrared gas
analyzer (ADC model LCA-2). Air was supplied and controlled by means of an ADC
model ASUM2 pump. This equipment was fitted to one of two custom-made
cuvettes, depending on whether current-year or 1-year-old foliage was being measured. In the 1989 experiment, only current-year foliage was measured. The cuvettes,
which were designed to enclose single shoots, had small volumes (< 100 cm3). On
completion of all tests, the shoots were cut and weighed. Needle dry weight was used
to standardize the measured gas exchange rates.
Temperatures within the cuvettes were controlled by decreasing the flow rate in the
chamber. Initial measurements were made at a flow rate of 500 l min −1, resulting in
a chamber air volume turnover rate of about six times per minute. This rate was
necessary to maintain the chamber temperature near ambient, and was probably a
result of the small chamber size (< 1 cm larger in diameter than the shoot). The
1342
VANN, JOHNSON AND CASPER
temperature was increased by reducing the flow to 100 l min −1. Slower flows resulted
in a temperature increase that was too rapid to allow determination of a stable
photosynthetic rate, and faster flows took too long to reach the desired temperatures.
Use of flow rate to adjust temperature was chosen for two reasons. First, it was more
convenient for field determinations, because it required no additional equipment.
Second, increased temperature can result in an increase in the water vapor pressure
gradient (Kramer 1983). Some of the effect of heat on photosynthesis may be a result
of desiccation caused by this increase in the vapor deficit (Larcher 1983). Reducing
the air flow permitted the chamber air to become saturated with water vapor from
transpiration, thus minimizing effects arising from desiccation. Although altering the
flow rates may be expected to alter the boundary layer size, calculations (based on
equations in Nobel 1991) suggest that the actual change was less than 10%. Temperature was determined by placing a microthermocouple against the underside of a
single needle in the upper portion of the enclosed shoot; all temperatures reported are
thus needle surface temperatures.
Once the chamber was installed, the twig was given about 10 min for gas exchange
to stabilize. Two readings were then taken 5 min apart for the determination of
photosynthetic rate at ambient temperature. The flow was then decreased and the
chamber was covered with a sleeve of black plastic and shaded. Five min later,
respiration rate was determined, and the shading and sleeve were removed. The flow
rate was held low until the chamber temperature reached about 40 °C. One measurement was taken after about 15--30 min had elapsed, when the temperature was in the
mid-30 °C range, and a second measurement was taken when the temperature had
reached about 40 °C. To obtain a duplicate measurement, the branch was held at
about 40 °C for 15 min by intermittently shading the chamber. The flow rate was then
increased to 1 l min −1 for 30 s to flush the chamber, then returned to the low flow rate
(needed to improve the resolution of the respiration rate measurement). The chamber
was then covered with a black plastic sleeve, and the temperature permitted to rise to
about 40 °C. As it was not possible to hold the chamber at a constant temperature for
more than a few minutes at a time, two measurements were taken 2 min apart to
determine the respiration rate at the elevated temperature.
In 1989, measurements were made on September 9; in 1990, current-year foliage
was measured on August 3 and 4, and 1-year-old foliage was tested 1 week later.
Because it was desirable to maintain consistent light conditions during the experiment, measurements were conducted only when the site was in direct sun, a period
of a few hours in the early afternoon. This kept the shoots under study in saturating
light conditions and in appropriate conditions for solar heating. Because it took about
20 min to install and check the chamber for leaks, and over an hour to run the test, it
was possible to analyze only two trees per day. Measurements on 1-year-old foliage
were obtained for only one branch per tree. On subsequent days, four of the
current-year twigs that had been tested the previous week were remeasured to
determine recovery.
During the 1989 experiment, the mountain had a light cloud cover, reducing
HEAT AND CARBON DIOXIDE EXCHANGE IN RED SPRUCE
1343
ambient irradiance to about 230 µmol m −2 s −1. Although this is well below the light
saturation value, it is adequate for measurable carbon gain in this species (Manley
and Ledig 1979). The 1990 experiment was conducted on cloudless days with an
irradiance of about 1800 µmol m −2 s −1, well above the light saturation point (Manley
and Ledig 1979, McLaughlin et al. 1990).
Results
In 1987, there was a loss of chlorophyll and carotene in the visibly injured cells of
red spruce needles (Figure 1). The change was most pronounced in the 1-year-old
(1986) needle tissue. Visibly injured 1-year-old needles lost 45% of their chlorophyll
compared with uninjured needles from nearby shoots (T ≈ 13.1, P < 0.002). Carotenoids also decreased significantly, but to a lesser degree than chlorophyll (37.8%, T
≈ 10, P < 0.006). The current-year (1987) needles also lost significant amounts of
chlorophyll (34.6%, T ≈ 11.9, P < 0.003) and carotenoids (26.5%, T ≈ 3.46, P < 0.03).
The loss of chlorophyll in chlorotic tissues resulted from the breakdown of
chloroplast membranes (Figures 2A and 2B). Figure 2D shows a chloroplast with a
normal membrane system taken from a green needle in the same chamber. In the
damaged needles, the thylakoids had disintegrated, leaving only lipid droplets (od)
(cf. Levitt 1980). In addition, the chloroplasts became more spherical in shape.
Injured cells also exhibited an altered precipitation of vacuolar contents (Figure 2C).
Although the relationship between vacuolar chemistry and the visible structural
differences is not understood, the visible changes were consistent in all cells examined, and may reflect disruptions of the cellular osmotic environment. In some cells,
cellular structures had precipitated completely, possibly as a result of tonoplast
rupture. Needles that were visibly damaged became increasingly necrotic during the
following two weeks, and eventually abscised.
Figure 1. Effect of elevated air temperature on pigment concentrations of current and 1-year-old needles
of red spruce sampled in 1987. U = uninjured needles, I = visibly injured, j = chlorophyll, and h =
carotenoids. Bars = 95% c.i.
1344
VANN, JOHNSON AND CASPER
Figure 2. Electron micrographs of heat-damaged (A--C) and undamaged needle cells (D--E). (A) cell,
(B) chloroplast detail, (C) vacuole detail, (D) chloroplast detail, and (E) cell. Legend: n = nucleus, m =
mitochondrion, cw = cell wall, sg = starch grain, og = osmiophilic globule, od = oil droplet, nvt = normal
vacuolar tannin, and pvt = precipitated vacuolar tannin.
There was a rapid decrease in photosynthetic rate at temperatures above 30 °C
(Figure 3). At temperatures around 40 °C, gross photosynthetic rates declined below
the CO2 compensation point. Photosynthetic rates in 1-year-old needles were about
30% lower than in current-year needles (Figure 3B). Figure 3A shows data for
current-year needles for both 1989 and 1990. The initial values below 20 °C were
from the trees measured in 1989 under overcast skies. The regression between the
temperature response (Tmp) and photosynthetic rate (Ps) in current-year needles had
a steeper slope in 1990 than in 1989. The equation for the 1989 current-year needles
was Ps = 7.25 − 0.225Tmp (R2 = 0.753, P < 0.01), and for the 1990 current-year
needles was Ps = 12.3 − 0.331Tmp (R2 = 0.952, P < 0.01). The response of 1-year-old
tissue was similar: Ps = 7.79 − 0.234Tmp (R2 = 0.925, P < 0.01). The low R2 value in
the 1989 current-year foliage presumably reflects differences between the trees; one
was declining, the other visibly healthy. The high R2 values for the trees measured in
1990 may be a consequence of selecting trees that were similar. All tissues declined
toward a common value of zero, reducing the variance at high temperatures.
Respiration approximately doubled over the temperature range of 10 to 15 °C. This
HEAT AND CARBON DIOXIDE EXCHANGE IN RED SPRUCE
1345
Figure 3. Effect of temperature on CO2 exchange in (A) current-year needles and (B) 1-year-old needles
of red spruce. Apparent photosynthesis at ambient temperatures = d, at 10 °C above ambient = e, and
at 15 °C above ambient = j. Respiration at ambient temperature = n, and at 15 °C above ambient = ,.
Shaded symbols denote data collected in 1989; other symbols denote data collected in 1990. Temperatures are needle surface temperatures.
measurement may be low, because respiration in this species is low, approaching the
sensitivity of the instrument used. To increase the ∆CO2 value, the respiration
measurements were determined at very low flow rates. This resulted in high chamber
humidities accompanied by some condensation in the exhaust tube, and may have
altered CO2 concentrations through stomatal closure or through dissolution into the
water droplets. Respiration rates increased to similarly high values in both years,
although the temperatures were not comparable. The regression between temperature
response (Tmp) and respiration rate (Rs) was performed individually on each year’s
data. The equations obtained for the current-year needles were Rs = 0.347 − 0.089Tmp
(R2 = 0.955, P < 0.01) and Rs = 1.30 − 0.089Tmp (R2 = 0.937, P < 0.01) for 1989 and
1990, respectively. The 1-year-old needles exhibited a more gradual response; the
equation was Rs = − 0.434 − 0.036Tmp (R2 = 0.68, P < 0.01). Although the rate of
response in 1-year-old needles was unexpectedly low, given that these tissues were
the first to become visibly damaged in response to overheating, the respiration rates
at the higher temperature were significantly different from those at the ambient
temperature (T = 3.84, P < 0.01). The measurements with 1-year-old needles covered
a smaller temperature range than the current-year tests, partly because of shading
from adjacent twigs. Respiration of the 1-year-old needles may have increased
nonlinearly at higher temperatures. Because the cuvettes were small and could only
accommodate a single shoot (typically 3--4.5 cm in length), and flow rate was low,
the boundary layer may have been relatively large, thereby reducing the accuracy of
1346
VANN, JOHNSON AND CASPER
the measurement.
The second set of measurements taken in 1990 on current-year foliage were 92 ±
7% of the values determined the week before (data not shown). These values were
taken to indicate complete recovery of the photosynthetic capacity from overheating;
some of the difference in rates was likely due to different weather conditions on the
day of measurement.
Discussion
Manipulation of chamber temperatures did not precisely mimic conditions in nature,
because the humidity in the chamber remained near saturation. In nature, the plant
would lose water vapor freely. This permits cooling, but under some conditions could
result in water stress, which would exacerbate the effects of heat on photosynthesis
(Levitt 1980, Kramer 1983, Kozlowski et al. 1991). Although transpirational cooling
may reduce heating, conifer needles typically track ambient temperatures when there
is wind (Tibbals et al. 1964), and red spruce needles in full sun with little wind may
be 3--4 °C warmer than the air (D. Vann, unpublished data). In addition, transpiration
rates are low in conifers, resulting in decreases in needle temperature due to evaporative cooling of only 1--3 °C (Tibbals et al. 1964). Therefore, it is not clear whether
there is sufficient transpirational cooling in this species to protect it from injury at
high temperatures. Trees at low elevations may often be exposed to injurious needle
temperatures, particularly in the southern Appalachians where air temperatures often
reach 35--37 °C (Environmental Data Services 1968). In this experiment, injuries
were presumed to have occurred primarily or wholly as a result of direct heating,
because the high humidity must have reduced transpirational cooling.
Visible chlorosis, indicating severe cellular injury, occurred between 32 and 40 °C.
This is about 20 °C lower than the range of 50--55 °C reported for heat shock injury
in temperate plants, and about 10 °C below values causing damage after long-term
exposure (see compilations in Levitt 1980, Kozlowski et al. 1991). The needle tissues
damaged were generally from trees that were declining, and may have been sensitized to additional stress.
Temperatures can frequently reach values that reduce the net carbon gain of the
plant, but do not cause tissue death. Above 30--35 °C, red spruce trees fix little, if
any, carbon. The temperature compensation point values found for red spruce in this
study are similar to those reported for other cold temperate conifers (Larcher 1969,
Fryer and Ledig 1972). It should be noted that the short duration of the heat treatment
in this experiment likely underestimated the actual effect of high temperatures on
carbon gain, because exposure times in nature might last several hours per day. The
decrease in CO2 uptake in response to elevated temperatures suggests that a reduction
in growth will occur if temperatures are elevated for long periods (cf. Levitt 1980).
This prediction is consistent with evidence from dendrochronological studies in red
spruce indicating that unusually warm August temperatures are followed by smaller
rings in the following year (Cook and Zedaker 1992).
HEAT AND CARBON DIOXIDE EXCHANGE IN RED SPRUCE
1347
Although linear regressions were used to evaluate the decline in Ps, this does not
demonstrate that the decline was linear with increasing temperature. Data were not
taken at a sufficient range of temperatures to establish the shape of the curve;
however, published data suggest that such curves are parabolic (Larcher 1969). The
higher photosynthetic rates at 25 °C may indicate an optimum, but may also be
related to the age (youth, vigor) of the trees measured in 1990 and cannot be regarded
as a continuation of those measured at 15 °C. For this reason, a parabolic curve was
not fitted to the whole data set, although it does appear to correspond with maxima
suggested by the data of Manley and Ledig (1979).
Current-year needle respiration rates at elevated temperatures were similar in both
years, although the maximum temperatures attained differed by almost 10 °C. This
may be due to an inherent maximum respiration rate due to enzymatic limitations, or
may represent a decrease at the higher (near 40 °C) temperature as a consequence of
the breakdown of cellular components. Alternately, it may be related to the different
trees used in the two experiments; compared with the trees measured in 1990, the
1989 trees were 75--100 years older, 270 m higher in elevation and, in one case,
declining. All of these factors may influence respiration rates. The elevational effect
may explain the significantly (T = 2.78, P < 0.02) higher respiration rates in the
high-elevation trees, which were measured at a lower ambient temperature than the
low-elevation trees. However, based on an extrapolation of the regression for temperature increase, and assuming the high-elevation trees had been measured at an
ambient temperature of 25 °C, respiration rates would still have been higher in the
high-elevation trees than in the low-elevation trees, suggesting that that high-elevation trees may be more sensitive to high temperatures than low-elevation trees.
We conclude that the effect of elevated temperature on red spruce is an important
consideration in explaining its geographical range, and suggest that red spruce’s
historic migrations may have been driven, at least in part, by the reduction in growth
and reproduction accompanying an unfavorable carbon balance at warm temperatures.
Modeling efforts based on past distributions reconstructed from the palynological
record and current ecology show that small increases in temperature may change the
distribution of red spruce dramatically (Davis and Botkin 1985). Based on studies
with a model designed to extrapolate the consequences of global climate warming,
Overpeck et al. (1991) concluded that during the next 2--5 centuries, red spruce
would be eliminated south of about latitude 45° N. Ecotypic variants for optimal
photosynthetic temperatures have been described in balsam fir (Fryer and Ledig
1972). Similar variants may exist in red spruce, because photosynthetic optima have
been shown to respond to prior conditions (Manley and Ledig 1979). Thus, the
species may adapt to changing climates; however, each ecotype has some genetic
limit to its adaptability. Therefore, the species must rely on reproduction to migrate
out of unfavorable habitats. Based on data in Randall (1974) and Blum (1990), the
migration rate of red spruce is estimated to be about 100 m year −1. Over the
period described by Overpeck et al. (1991), red spruce may move 20--50 km. Even
1348
VANN, JOHNSON AND CASPER
assuming that this estimate is conservative, only the northernmost populations of red
spruce would survive. Thus, susceptibility to high temperatures combined with a low
migration rate may make red spruce an early victim of global warming.
Acknowledgments
We thank Scott D. Russell of the University of Oklahoma and his staff for the electron micrographs. We
also thank Julian L. Hadley for helpful comments and review of the manuscript. This research was funded
in part by a grant from the Andrew Mellon Foundation.
References
Blum, B.B. 1990. Picea rubens Sarg. Red spruce. In Silvics of North America: 1. Conifers. Agric.
Handbook No. 674. Eds. R.M. Burns and B.H. Honkala. USDA Forest Service, Washington, D.C.,
pp 250--259.
Cook, E.R. and S.M. Zedaker. 1992. The dendroecology of red spruce decline. In Ecology and Decline
of Red Spruce in the Eastern United States. Eds. C. Eager and M.B. Adams. Springer-Verlag, New
York, pp 192--231.
Davis, M.B. and D.B. Botkin. 1985. Sensitivity of cool temperate forests and their fossil pollen to rapid
climatic change. Quat. Res. 23:327--340.
Davis, M.B., R.W. Spear and L.C.K. Shane. 1980. Holocene climate of New England. Quat. Res.
14:240--250.
Environmental Data Services. 1968. Climatic atlas of the United States. US Dept. Commerce, Washington, D.C., 80 p.
Fryer, J.H. and F.T. Ledig. 1972. Microevolution of the photosynthetic temperature optimum in relation
to the elevational complex gradient. Can. J. Bot. 50:1231--1235.
Hayat, M.A. 1981. Principles and techniques of electron microscopy. University Park Press, Baltimore,
522 p.
Ibe, R.A. 1985. Postglacial montane vegetational history around Balsam Lake, Catskill Mountains, New
York. Bull. Torr. Bot. 112:176--186.
Kozlowski, T.T., P.J. Kramer and S.G. Pallardy. 1991. The physiological ecology of woody plants. In
Physiological Ecology Series. Ed. H.A. Mooney. Academic Press, San Diego, CA, 657 p.
Kramer, P.J. 1983. Water relations of plants. Academic Press, Orlando, FL, 489 p.
Larcher, W. 1969. The effect of environmental and physiological variables on the carbon dioxide
exchange of trees. Photosynthetica 3:167--198.
Larcher, W. 1983. Physiological plant ecology. Springer-Verlag, Berlin, 303 p.
Levitt, J. 1980. Responses of plants to environmental stresses: chilling, freezing, and high temperature
stresses. In Physiological Ecology Series. Ed. J. Levitt. Academic Press, Orlando, FL, 497 p.
Manley, S.A.M. and F.T. Ledig. 1979. Photosynthesis in black and red spruce and their hybrid derivatives:
ecological isolation and hybrid adaptive inferiority. Can. J. Bot. 57:305--314.
Mark, A.F. 1958. The ecology of the southern Appalachian grass balds. Ecol. Monogr. 28:293--336.
McLaughlin, S.B., C.P. Andersen, N.T. Edwards, W.K. Roy and P.A. Layton. 1990. Seasonal patterns of
photosynthesis and respiration of red spruce saplings from two elevations in declining southern
Appalachian stands. Can. J. For. Res. 20:485--495.
Müller, W.C. and E. Rodehurst. 1977. The effect of some alkaloids on the ultrastructure of phenolic
containing cells in the endodermis of cotton roots. In Proc. 35th Annual EMSA Meetings, Baton
Rouge, LA, 544 p.
Nobel, P.S. 1991. Physicochemical and environmental plant physiology. Academic Press, San Diego,
CA, pp 396--398.
Overpeck, J.T., P.J. Bartlein and T. Webb, III. 1991. Potential magnitude of future vegetation change in
eastern North America: comparisons with the past. Science 254:692--695.
Randall, A.G. 1974. Seed dispersal into two spruce--fir clear-cuts in eastern Maine. Univ. Maine Life Sci.
Agric. Exp. Stn. Res. Life Sci. 21:15.
HEAT AND CARBON DIOXIDE EXCHANGE IN RED SPRUCE
1349
Salerma, J.T. and M.R. Brandaõ. 1973. The use of PIPES buffer in the fixation of plant cells for electron
microscopy. J. Submicroscop. Cytol. 5:79--83.
Spurr, A.R. 1969. A low-viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastruct. Res. 26:31--42.
Tibbals, E.C., E.K. Carr, D.M. Gates and F. Kreith. 1964. Radiation and convection in conifers. Am. J.
Bot. 51:529--538.
Vann, D.R. and A.H. Johnson. 1988. Design and testing of a field branch enclosure for the exclusion of
atmospheric components. In Proc. on Effects of Atmospheric Pollutants on the Spruce--Fir Forests of
the Eastern United States and the Federal Republic of Germany. USDA Forest Service, Northeast
Forest Exp. Stn., Broomall (Radnor), PA, pp 431--439.
Vann, D.R., G.R. Strimbeck and A.H. Johnson. 1992. Effects of ambient levels of airborne chemicals on
freezing resistance of red spruce foliage. For. Ecol. Manage. 51:69--79.
Watts, W.A. 1979. Late Quarternary vegetation of central Appalachia and the New Jersey coastal plain.
Ecol. Monogr. 49:427--469.
Wellburn, A.H. and H. Lichtenthaler. 1984. Formulae and program to determine total carotenoids and
chlorophylls a and b of leaf extracts in different solvents. In Advances in Photosynthesis Research,
Vol. 2. Ed. C. Sybesma. Martinus Nijhoff/Dr. Junk Publishers, The Hague, Netherlands, pp 9--12.
Whitehead, D.R. 1973. Late-Wisconsin vegetational changes in unglaciated eastern North America.
Quat. Res. 3:621--631.
Whittaker, R.H. 1956. Vegetation of the Great Smoky Mountains. Ecol. Monogr. 26:1--80.