Productivity responses of Acer rubrum and Taxodium distichum
seedlings to elevated CO2 and ?ooding
C.D. Vanna,c,*, J.P. Megonigalb
aUS Department of the Interior, US Geological Survey, MS 926A, Reston, VA 20192, USA
bSmithsonian Environmental Research Center, Edgewater, MD 21037, USA
cGeorge Mason University, Fairfax, VA 22030, USA
??Capsule??: The growth enhancing e?ects of elevated atmospheric carbon dioxide were not observed in young seedlings of
two tree species in ?ooded soils due to soil oxygen limitations.
Abstract
Elevated levels of atmospheric CO2 are expected to increase photosynthetic rates of C3 tree species, but it is uncertain whether
this will result in an increase in wetland seedling productivity. Separate short-term experiments (12 and 17 weeks) were performed
on two wetland tree species, Taxodium distichum and Acer rubrum, to determine if elevated CO2 would in?uence the biomass
responses of seedlings to ?ooding. T. distichum were grown in replicate glasshouses (n=2) at CO2 concentrations of 350 or 700
ppm, and A. rubrum were grown in growth chambers at CO2 concentrations of 422 or 722 ppm. Both species were grown from seed.
The elevated CO2 treatment was crossed with two water table treatments, ?ooded and non-?ooded. Elevated CO2 increased leaf-
level photosynthesis, whole-plant photosynthesis, and trunk diameter of T. distichum in both ?ooding treatments, but did not
increase biomass of T. distichum or A. rubrum. Flooding severely reduced biomass, height, and leaf area of both T. distichum and A.
rubrum. Our results suggest that the absence of a CO2-induced increase in growth may have been due to an O2 limitation on root
production even though there was a relatively deep (
10 cm) aerobic soil surface in the non-?ooded treatment. # 2001 Elsevier
Science Ltd. All rights reserved.
Keywords: Elevated CO2; Wetlands; Tree seedlings; Productivity; Taxodium distichum; Acer rubrum
1. Introduction
Since the industrial revolution, atmospheric CO2
concentrations have increased by 30% to 360 ppm
(Keeling and Whorf, 1994) and are predicted to rise to
500 ppm by the year 2050 (Kattenburg et al., 1995).
Most upland tree species respond to increased atmos-
pheric CO2 concentrations by increasing photosynthetic
rate, which generally results in increased productivity
(Curtis and Wang, 1998; Delucia et al., 1998; Norby et
al., 1999). However, no studies have been conducted on
the productivity responses of wetland tree species to
elevated CO2. Such responses may in?uence the full
range of ecological processes that operate in wetland
systems, including seedling establishment and survival.
Flooding is the most important factor limiting tree seed-
ling survival in forested wetlands. Survival of wetland tree
seedlings is generally poor during the ?rst growing season,
then increases substantially in subsequent seasons as meta-
bolic and morphological adjustments occur (Streng et al.,
1989; Megonigal and Day, 1992; Jones et al., 1994).
Because ?ood tolerance in wetland tree seedlings
increases as they develop, any factor a?ecting growth
rate may potentially improve seedling survival.
Anoxic conditions incur energetic (carbon) costs on
plants for production of specialized morphological
structures such as aerenchyma tissue, adventitious roots
and lenticels (Hook, 1984; Kozlowski, 1984; Day, 1987;
Mitsch et al., 1991). It also limits plant growth through
ine?cient carbon metabolism during fermentation
respiration (Crawford, 1983) and accumulation of soil
toxins such as hydrogen sul?de (Mendelssohn et al.,
1981). The growth responses of wetland seedlings to
elevated CO2 may be restricted if O2-de?cient soils cause
a growth?sink limitation. If so, elevated CO2 would only
improve seedling growth under relatively dry conditions.
Our objectives were to determine the growth responses
of wetland tree seedlings to elevated CO2 and water
table depth. The two species investigated, Taxodium
distichum and Acer rubrum, di?er in their potential to
adapt to ?ooding stress. We hypothesized that elevated
CO2 would increase productivity of both species in the
non-?ooded treatment, but Acer rubrum would not
0269-7491/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PI I : S0269-7491(01 )00244-5
Environmental Pollution 116 (2002) S31?S36
www.elsevier.com/locate/envpol
* Corresponding author. Fax: +1-703-648-6953.
E-mail address: cdvann@usgs.gov (C.D. Vann).
respond to elevated CO2 in the ?ooded treatment due to
an O2 limitation.
2. Materials and methods
Separate experiments were conducted during the
summer of 1997 at the Smithsonian Environmental
Research Center (SERC, Edgewater, MD) and 1998
at the Duke University Phytotron facility (Durham,
NC). The experimental design and growth conditions
used in the two studies are compared in Table 1. The
two tree species used in these experiments di?ered in
their ability to tolerate ?ooding. Acer rubrum is a
deciduous broad-leaved tree that can tolerate seasonal
inundation for only 25% of the growing season
(Thurnhorst and Biggs, 1993). Taxodium distichum is a
deciduous woody conifer that can tolerate hydrologic
regimes ranging from well-drained to 3 m of ?ooding
for the entire growing season (Mattoon,1915).
2.1. Experimental procedure?Acer rubrum
Acer rubrum seedlings were germinated in a mixture
of 50% Canadian Sphagnum peat and 50% vermiculite/
reed sedge peat (Ho?man Professional Potting mix,
Lancaster, PA) that was kept moist but not ?ooded.
Upon germination the water depth was increased to the
soil surface. Seeds were germinated under ambient CO2
conditions, ?ltered sunlight and temperatures of
approximately 26C/22C (day/night).
Seedlings of approximately equal size (6 cm in height)
with six true leaves were transplanted on 18 June 1997
into 10 cm-diameter38 cm polyvinyl chloride (PVC)
containers with 10 cm-diameter PVC end-caps on the
base (Table 1). There was one transplanted seedling per
PVC container. Two 350.3 cm-vertical slits allowed
water to ?ow into the pots and replace transpirational
losses. A strip of aluminum screen secured over the slits
inside the container prevented excessive soil loss. The
pots were ?lled to within 5 cm of the top with a soil mix
prepared 1.5 months prior to transplantation (Table 1).
The soil mix and pH approximated tidal freshwater
forested wetland soils on the White Oak River, NC
where A. rubrum and T. distichum co-occur (Megonigal,
1996). A slow release all-purpose fertilizer was mixed
into the soil prior to transplantation and this was sup-
plemented with the addition of a water-soluble fertilizer
applied once per month at half strength beginning on
18 June. The transplanted seedlings were grown from 18
July to 8 November 1997 in four chambers (2 m length
0.9 m width1.5 m height) located in a glasshouse at
the SERC facility. CO2 concentrations were maintained
near 422 and 722 ppm for each replicate (n=2) growth
chamber.
Individual seedlings were randomly assigned to CO2
and water treatments. Twelve containers were placed
into 133-l plastic tubs. The soil surface in half of the
containers was ?ooded and the other half was elevated
above the water table by setting them on bricks
(Table 1). There were four no-plant controls per treatment.
Water levels for transplanted seedlings were initially set
Table 1
Growth parameters for the SERC and Duke Phytotron studies
Study
SERC Facility 1997 Duke Phytotron 1998
Species Acer rubrum Taxodium distichum
Facility Growth chamber Glasshouse
Replicate units 2 2
Replicate pots in
each unit unit
Six per water table treatment Eight per water table treatment
Treatments CO2, Water CO2, Water
CO2 levels Ambient?422 ppm CO2; Elevated 722 ppm CO2 Ambient?350 ppm CO2; Elevated 700 ppm CO2
H2O levels +2 cm ?ooded; 9 cm non-?ooded +5 cm ?ooded; 10 cm non-?ooded
Soil 92% Canadian Sphagnum peat 95% Canadian Sphagnum peat
6% muck soil from White Oak River, NC 3% muck soil from White Oak River, NC
2% pulverized lime 2% dolomitic lime
Pot diameter 10 cm 10 cm
Pot depth 38 cm 43 cm
Duration of study 17 weeks 12 weeks
Light Filtered Sunlight (1200 mmol m2 s1 at 13:00 h) Filtered Sunlight (1500 mmol m2 s1 at 13:00 h)
Temperature 24 ?29C Day/21?25C Night 28C (06:00?20:00 h) and 23C overnight
Water temperature 24?30C Day/23?27C Night N/A
Fertilizers Osmocote 18-6-12 NPK (rate?81 ml/0.03 m3 soil Scott?s Master Collection 15-13-13 NPK (rate ? 95 ml/0.03 m3)
Miracle-Gro 15-30-15 NPK (rate? 50% strength/3 weeks) Miracle-Gro 15-30-15 NPK (rate? 50% strength/3 weeks)
Harvest intervals 17 weeks 12 weeks
Plant source Seed, Environmental Concern (St. Michaels, MD) Seed, F.W. Schumacher Co., Inc. (Sandwich, MA)
Photosynthesis N/A Leaf-level and whole-plant
S32 C.D. Vann, J.P. Megonigal / Environmental Pollution 116 (2002) S31?S36
at the soil surface for both the ?ooded and non-?ooded
treatments on 18 June. As the seedlings grew, water
levels were slowly changed over a 49-day period to +2
cm (above the soil surface) for the ?ooded treatment
and 9 cm (below the soil surface) for the non-?ooded
treatment (also referred to as saturated). Water levels
were maintained at the target levels by replacing evapo-
transpiration losses with tap water on a daily basis.
Styrofoam ??peanuts?? were placed on the surface of the
water to minimize algal growth and evaporation. Tubs
were drained and water replaced once per month to pre-
vent salt accumulation. Containers were rotated 3 times
per week within each tub to mitigate tub location e?ects.
Total biomass was determined by destructive har-
vesting. Tree height, leaf area and total number of
branch sets were measured. Roots were gently washed
of soil and separated by hand. Plant material was dried
for 5 days at 70 C and weighed.
2.2. Experimental procedure?Taxodium distichum
In a separate experiment, T. distichum seeds were
stored for 6 days at 5C under dry conditions, and then
soaked for 48 h in 0.01% nitric acid to break dormancy.
They were planted into ?ats ?lled with Canadian
Sphagnum peat moss, amended with dolomitic lime to a
pH of approximately 6.7. Germination ?ats were placed
into two separate environmentally controlled growth
chambers at the Duke University Phytotron in which
CO2 concentrations were maintained near 350 or 700
ppm. Seedling germination occurred at CO2 concentra-
tions identical to those used in the glasshouse treat-
ments. Temperatures were 25C/20C (day/night) and
photon ?ux density was 1000 mmol m2 s1 during a 14-
h day-length period. Relative humidity in both cham-
bers was >70%. Soils were maintained in a moist, but
non-?ooded condition until transplantation because T.
distichum seeds continued to germinate over a time period
of 
3 weeks, whereas all of the A. rubrum seeds had
germinated within a week. T. distichum seeds require a
moist, but non-?ooded soil to germinate.
Plants of approximately equal size with six true leaves
(T. distichum) were transplanted on 23 June 1998 into 10
cm-diameter43 cm PVC containers with 10 cm-dia-
meter PVC end-caps on the bases. There was one
transplanted seedling per PVC container. Each con-
tainer had sets of four holes (about 1 cm-diameter) at 6
cm and at 28 cm above the base to permit water
exchange. A strip of aluminum screen was secured over
each set of holes to prevent excessive soil loss. The bot-
toms of the container had 250 ml of pea gravel to pre-
vent it from ?oating. Containers were ?lled to within 4
cm of the top with a peat-based soil mixture (similar to
that used for A. rubrum) 3 weeks before transplanting.
A slow release all-purpose fertilizer was mixed into the
soil prior to transplantation and this was supplemented
with the addition of a water-soluble fertilizer applied
every 3 weeks at half strength beginning on 20 June. The
transplanted seedlings were grown in four environmen-
tally controlled glasshouses at the Duke University
Phytotron from 23 June to 16 September 1998 (Table 1).
Individual seedlings, within each CO2 treatment (350
or 700 ppm), were randomly assigned to water table
treatments. The water table was manipulated in 83-l
plastic tubs. There were two tubs per species in each
greenhouse, one ?ooded and one not ?ooded. There
were eight no-plant controls per treatment. Styrofoam
disks (0.6 cm thick) were placed on the surface of the
water in each tub to minimize algal growth and evapo-
ration. Water levels were initially set at 6 cm for the
non-?ooded and 2 cm for the ?ooded water treat-
ments, and then changed to 10 cm for the non-?ooded
treatment and +5 cm for the ?ooded treatment over an
11-day period. Water levels were maintained at the tar-
get levels by replacing evapotranspiration losses with
tap water on a daily basis. Tubs were drained and re?l-
led every 3 weeks to prevent salt accumulation.
Total plant biomass was determined by destructive
harvesting. Shoot height, leaf area and trunk diameter
were measured. Roots were gently washed of soil and
separated into two categories, ?ne4 0.1 cm) and coarse
(>0.1 cm). Plant material was dried for 5 days at 70C
and weighed.
2.3. Leaf-level and whole-plant photosynthesis
In situ photosynthetic measurements were made on a
subset of 6 replicate T. distichum plants per treatment
block using a Licor model 6200 portable photosynthetic
system (Lincoln, NB) equipped with a 0.25-l leaf cham-
ber. Glasshouse photon ?ux densities ranged from 1300
to 1500 mmol m2 s1 during the measurement period
(09:30 to 16:00 h). Photosynthetic measurements were
made at 
5 day intervals between 11 July and 15 Sep-
tember on leaves that were both new (within three
branches of the apex) and old (within three branches of
the base). At the time of harvest, individual leaf ages
and leaf areas (determined using a Leaf Area Meter)
were used to calculate whole-plant photosynthesis:
Whole-plant photosynthesis
?
Xn
i?1
LAold Aold? ? ? LAnew Anew? ?
where LA is the total area of leaves and A is the age-
speci?c photosynthesis.
2.4. Statistical analysis
The SAS univariate procedure was used to assess nor-
mality (SAS Institute, 1987) and Levene?s Test for
C.D. Vann, J.P. Megonigal / Environmental Pollution 116 (2002) S31?S36 S33
Equality of Variance was used to assess homoscedascity.
Data were log transformed when variances among
treatments were unequal. Data were analyzed by a two-
way ANOVA with CO2 and H2O as main e?ects; CO2
e?ects were calculated using a Type III mean-square-
error term with glasshouse nested within the CO2 main
e?ect.
3. Results and discussion
Elevated CO2 signi?cantly increased leaf-level and
whole-plant photosynthesis of T. distichum seedlings
regardless of water treatment (P<0.05, Fig. 1). The
increase in whole-plant photosynthesis ranged from 58
to 128% (replicate glasshouses averaged). Flooding sig-
ni?cantly reduced leaf-level photosynthesis of new
leaves and whole-plant photosynthesis (P<0.01). There
were no water treatment e?ects on leaf-level photo-
synthesis of old T. distichum leaves suggesting a degree
of acclimation to ?ooding in old leaves, compared with
new leaves. There were no CO2H2O interactions for
either leaf-level or whole-plant photosynthesis. Photo-
synthetic rates of A. rubrum were not measured.
Although T. distichum photosynthetic rates increased
in response to a CO2-enriched atmosphere, elevated
CO2 did not signi?cantly increase T. distichum biomass
after 12 weeks of treatment (Fig. 2). Likewise, A.
rubrum did not increase in biomass after 17 weeks of
treatment. The absence of an increase in biomass sug-
gests that some factor was limiting the conversion of
photosynthates to biomass. In addition, the lack of an
increase in A. rubrum biomass could have been due to
Fig. 1. Leaf photosynthesis and whole-plant photosynthesis of Taxodium distichum.Values are means
1 standard error. Each pair of bars repre-
sents the two replicate chambers or glasshouses. Signi?cant di?erences between pairs of bars are indicated by the letters C and W placed above the
pair with the higher mean (C, signi?cant CO2 e?ect; W, signi?cant water treatment e?ect).
Fig. 2. Dry mass of Acer rubrum and Taxodium distichum seedlings.
Values are means
1 standard error. Each pair of bars represents the
two replicate chambers or glasshouses. Signi?cant di?erences between
pairs of bars with di?erent ?ooding treatments are indicated by the
letter W placed above the pair with the higher mean. There were no
signi?cant CO2 e?ects.
S34 C.D. Vann, J.P. Megonigal / Environmental Pollution 116 (2002) S31?S36
an absence of an increase in photosynthesis. Studies on
upland trees have identi?ed nutrient limitation and
rooting volume as factors that can limit the biomass
response of seedlings to elevated CO2. It is unlikely
these factors limited productivity in the present study
because the seedlings were fertilized and roots occupied
a relatively small portion of the pot volume upon
harvest.
We suggest that the lack of a productivity response to
elevated CO2 in the ?ooded treatment was caused by O2
limitation operating on the root system, which imposed
a growth?sink limitation on plant productivity. Flood-
ing clearly limited overall productivity in our study,
signi?cantly decreasing T. distichum and A. rubrum
seedling biomass, height, number of branches (A.
rubrum only), trunk diameter and leaf area (P<0.05).
Seedlings also exhibited red leaf coloration indicating
stress. Several previous studies of T. distichum and A.
rubrum have also reported signi?cant reductions in
shoot and root dry mass in response to anaerobic soil
conditions (Dickson and Broyer, 1972; McLeod et al.,
1986; Day, 1987; Pezeshki et al., 1996). At least in
the case of T. distichum, the excess metabolic energy
a?orded by a CO2-induced increase in photosynthesis
was insu?cient to ameliorate the ?ooding stress.
The lack of a productivity response to elevated CO2 in
T. distichum after 12 weeks of treatment was a transient
e?ect due to age. When T. distichum seedlings were
permitted to grow for another 6 weeks in the non-
?ooded water treatment there was a signi?cant increase
in root biomass and a non-signi?cant trend (P<0.10)
towards an increase in shoot biomass (Vann, 2000).
Because elevated CO2 signi?cantly increased T. dis-
tichum trunk diameter (P<0.01, Table 2) but not height
or biomass, it seems that CO2-enrichment decreased
stem wood density. This e?ect would be consistent with
an increase in the volume of aerenchyma tissue and an
improved supply of O2 to the root system. Thus, the
productivity response of T. distichum to elevated CO2
observed after 18 weeks (Vann, 2000) appeared to fol-
low the development of morphological adaptations to
?ooding observed 6 weeks earlier as reported here.
Seedling ?ood tolerance develops somewhat slowly
because it requires morphological changes such as the
production of water roots and aerenchyma tissue
(Crawford, 1983; Hook, 1984). For example, T. dis-
tichum seedlings were more productive in a periodically
?ooded than a continuously ?ooded hydrologic regime
in their ?rst year of growth, but this di?erence was
negligible after three growing seasons (Megonigal and
Day, 1992). The narrowing of the biomass di?erence
between the two treatments coincided with development
of morphological adaptations to ?ooding (Megonigal
and Day, 1992).
Because photosynthesis of A. rubrum was not meas-
ured, it is possible that soil ?ooding prevented a CO2-
induced increase in photosynthesis. If photosynthesis
did increase in response to elevated CO2, then A. rubrum
seedlings did not exhibit a growth response even after 17
weeks of treatment. This species may not respond to
elevated CO2 under ?ooded or partially saturated (9
cm water table) conditions even as a mature tree
because of its modest ?ood tolerance.
In summary, elevated CO2 did not increase T. dis-
tichum or A. rubrum biomass in either of the water
treatment after 12 and 17 weeks of CO2 treatment,
respectively. The absence of a CO2 response was most
likely due to O2-limitation, which is particularly stress-
ful in the seedling stage. These results suggest that
the growth of wetland trees will not be in?uenced by
elevated CO2 during the ?rst several weeks following
germination.
Table 2
Morphological responses of Taxodium distichum and Acer rubrum to elevated CO2. Signi?cant di?erences are coded as follows; C, signi?cant CO2
e?ect and W, signi?cant water treatment e?ect
Species Flooded treatment Non-?ooded treatment Statistical
signi?cance
Ambient CO2 Elevated CO2 Ambient CO2 Elevated CO2
350 350 700 700 350 350 700 700
Taxodium distichum
Height (cm) 43.7 43.1 35.4 45.1 54.0 52.6 51.8 57.0 W
Leaf area (cm2) 152.0 134.9 95.1 132.8 276.5 295.6 235.0 272.3 W
Trunk diameter (cm) 0.40 0.41 0.38 0.43 0.43 0.42 0.49 0.43 CW
Ambient CO2 Elevated CO2 Ambient CO2 Elevated CO2
422 422 722 722 422 422 722 722
Acer rubrum
Height (cm) 10.8 8.2 11.1 9.1 17.6 15.6 20.4 19.5 W
Leaf area (cm2) 38.8 22.2 73.9 31.6 247.8 192.1 271.8 266.1 W
No. of branches 8.5 6.3 9.1 6.7 13.5 9.5 13.5 13.5 W
C.D. Vann, J.P. Megonigal / Environmental Pollution 116 (2002) S31?S36 S35
However, the results are not likely to apply to older
saplings that are more developed with respect to ?ood
tolerance. Additional studies are needed to establish the
potential for elevated CO2 to in?uence seedling estab-
lishment in wetland forests.
Acknowledgements
We thank Drs. R. Sharitz, F. Day and T. Cronin for
reviewing the manuscript and making helpful comments.
This research was supported by a National Science
Foundation grant (DEB-94-15541) awarded to the Duke
University Phytotron facility and a Department of Energy
grant (DOE-98-59-MP-4) awarded to J.P.M. Additional
funding was provided by the Summer Research Funding
Program, GeorgeMason University, VA. This paper was
presented at the USDA Forest Service Southern Global
Change Program sponsored Advances in Terrestrial
Ecosystem: Carbon Inventory, Measurements, and
Monitoring Conference held 3?5 October 2000 in
Raleigh, North Carolina.
References
Crawford, R.M.M., 1983. Root survival in ?ooded soils. In: Gore,
A.J.P. (Ed.), Ecosystems of the World. Elsevier Scienti?c Publish-
ing, Amsterdam, pp. 257?283.
Curtis, P.S., Wang, X., 1998. Ameta-analysis of elevated CO2 e?ects on
woody plant mass, form, and physiology. Oecologia 113, 299?313.
Day, F.P., 1987. E?ects of ?ooding and nutrient enrichment on bio-
mass allocation in Acer rubrum seedlings. American Journal of
Botany 74, 1541?1554.
Delucia, E., Hamilton, J., Naidu, S., Thomas, R., Andrews, J., Finzi,
A., Lavine, M., Matamala, R., Mohan, J., Hendry, G., Schlesinger,
W., 1999. Net primary production of a forest ecosystem with
experimental CO2 enrichment. Science 284, 1177?1179.
Dickson, R.E., Broyer, T.C., 1972. E?ects of aeration, water supply,
and nitrogen source on growth and development of tupelo, gum and
bald cypress. Ecology 53, 626?634.
Hook, D.D., 1984. Adaptions to ?ooding with fresh water. In:
Kozlowski, T.T. (Ed.), Flooding and Plant Growth. Academic
Press, Orlando, FL, pp. 265?294.
Jones, R.H., Sharitz, R.R., Dixon, P.M., Segal, D.S., Schneider, R.L.,
1994. Woody plant regeneration in four ?oodplain forests. Ecologi-
cal Monographs 64, 345?367.
Kattenburg, A., Giorgi, F., Grassl, H., Meehl, G.A., Mitchell,
J.B.F., Stou?er, R.J., Tokioka, T., Weaver, A.J., Wigley,
T.M.L., 1995. Climate models ? projections of future climate. In:
Houghton, J.T., Meira Fiho, L.G., Callander, B.A., Harris, N.,
Kattenburg, A., Maskell, K. (Eds.), Intergovernmental Panel on
Climate Change. Cambridge University Press, New York, pp.
290?349.
Keeling, C.D., Whorf, T.P., 1994. Atmospheric CO2 from sites in the
SIO air sampling network. In: Boden, T.A., Kaiser, D.P.,
Sepanski, R.J., Stoss, F.W. (Eds.), Trends 93: A Compendium of
Data on Global Change. Carbon Dioxide Information Analysis
Center, Oak Ridge National Laboratory, Oak Ridge, TN, pp. 16?
26.
Kozlowski, T.T., 1984. Responses of woody plants to ?ooding. In:
Kozlowski, T.T. (Ed.), Flooding and Plant Growth. Academic
Press, Orlando, FL, p. 129163.
Mattoon, W.R., 1915. The Southern Cypress. US Department of
Agriculture Bulletin 272.
McLeod, K.W., Donovan, N.J., Stump?, N.J., Sherrod, K.C., 1986.
Biomass, photosynthesis and water use e?ciency of woody swamp
species subjected to ?ooding and elevated water temperatures. Tree
Physiology 2, 341?346.
Megonigal, J.P., 1996. Methane production and oxidation in a future
climate. Ph.D. dissertation, Duke University, Durham, NC.
Megonigal, J.P., Day, F.P., 1992. E?ects of ?ooding on root and
shoot production of bald cypress in large experimental enclosures.
Ecology 73, 1182?1193.
Mendelssohn, I.A., McKee, K.L., Patrick Jr., R.H., 1981. Oxygen
de?ciency in Spartina alternifolia roots: metabolic adaptions to
anoxia. Science 214, 439?441.
Mitsch, W.J., Taylor, J.R., Benson, K.B., 1991. Estimating primary
productivity of forested wetland communities in di?erent hydrologic
landscapes. Landscape Ecology 5, 75?92.
Norby, R.J., Wullschleger, S.D., Gunderson, C.A., Johnson, D.W.,
Ceuemans, R., 1999. Tree responses to rising CO2 in ?eld experi-
ments: implications for the future forest. Plant, Cell and Environ-
ment 22, 683?714.
Pezeshki, S.R., Pardue, J.H., DeLaune, R.D., 1996. Leaf gas exchange
and growth of ?ood-tolerant and ?ood-sensitive tree species under
low soil redox conditions. Tree Physiology 16, 453?458.
SAS Institute, 1987. SAS/STAT Guide for Personal Computers, Ver-
sion 6 Edition. SAS Institute, Cary, NC.
Streng, D.R., Glitzenstein, S., Harcombe, P.A., 1989. Woody seedling
dynamics in an east texas ?oodplain forest. Ecological Monographs
59, 177?204.
Thurnhorst, G., 1993. Wetland Planting Guide for the Northeastern
United States.
Vann, C.D., 2000. The response of a woody plant-soil system to
elevated CO2 and ?ooding: carbon allocation to production and
methanogenesis. Master?s thesis, George Mason University, Fair-
fax, VA.
S36 C.D. Vann, J.P. Megonigal / Environmental Pollution 116 (2002) S31?S36