EPA/600/A-96/077
ENVIRONMENTAL MODIFICATION AND SHOOT GROWTH IN A
CLOSED ECOSYSTEM TO EVALUATE LONG-TERM RESPONSES OF
TREE SEEDLINGS TO STRESS
D.M. Olszyk and D.T. Tingey
U.S. EPA, National Health and Ecological Effects Laboratory, Western Ecology
Division, Corvallis, Oregon 97333
Abstract
Determination of ecosystem responses to stress requires careful control of the
environment and measurement of biological effects. Closed chambers provide
appropriate environmental control and measurement, but chamber impacts on the
ecosystems must be documented so that treatment effects can be interpreted properly. A
set of sun-lit chambers with control of atmospheric CO2, air temperature, and dew point
has been in operation for over 2 1/2 years. The chambers are being used to evaluate
responses of Douglas-fir (Pseudotsuga menziesii) seedlings to elevated CO2 and
temperature. Comparison of trees grown in chambers at ambient CO2 and temperature
(ACAT) with corresponding outside "chamberless trees" (CL) indicates that ACAT trees
have slightly less growth (e.g. smaller stem diameters, fewer branches, and shorter
terminal buds) compared to CL trees. However, needle characteristics are essentially the
same for ACAT and CL trees. The differences in growth for ACAT vs. CL trees likely
can be attributed to less soil moisture and lower light intensities, and possibly slightly
greater vapor pressure deficits in chambers vs. outside. Air temperature and CO2 levels
were very similar for ACAT and CL trees and likely do not affect tree growth. Thus,
growth of the trees to CO2 and temperature in the chambers are representative of the
imposed climatic conditions, and can be used to estimate effects of climate stress.
Keywords: Controlled environment, vegetation, Douglas-fir, Pseudotsuga menziesii
1. Introduction
Careful control of environmental conditions is essential for precise description of
factors affecting plant growth, and manipulation of climatic variables to test experimental
hypotheses. Many types of facilities have been used in plant-environment studies (Allen
et al., 1992) with closed environment chambers providing a high level of environmental
control, especially in terms of atmospheric gases, air temperature, and dew point (Taylor
et al., 1994). Sun-lit chambers (soil-plant-atmospheric-research or SPAR unit) has been
intensively developed to provide for CO2, air temperature, and dew point control (Jones
et al., 1984), as well as a controlled soil system through use of an attached lysimeter.
This type of chamber has been adapted at Corvallis, Oregon, for use with tree seedlings
and their associated soil ecosystem (Tingey et al., 1996).
This paper describes shoot responses of Douglas-fir seedlings after >2 1/2 years of
growth in the Oregon closed chambers with an ambient CO 2, temperature and dew point,
vj. seedlings grown in outside air, to indicate if the performance of the chambers is
providing conditions for acceptable plant growth. The trees are being grown as part of a
study on the effects of global climatic change (elevated CO2 and/or temperature) on
Douglas-fir and the soil-rhizosphere ecosystem (Tingey et al., 1996).
2. Methods
2.1 Chamber Operation

-------
The sun-lit, closed chambers ("terracosms") and associated equipment measure and
controlclimatic and edaphic factors while maintaining natural environmental variability as
described in detail by Tingey et al. (1996). Chambers are aligned on an east-west axis
with front sides facing south. In brief, each chamber has a top enclosure 2 m wide, 1 m
front-to-back, and 1.5 m tall in the back sloping to 1.3 m in the front, with an aluminum
frame covered with 3 mil thick Teflon™ film (Du Pont Electronics, Wilmington, DE) on
three sides and the top. The back (north) wall of the chambers is plexiglass. The back of
each enclosure has "air handling" equipment to modify CO2 concentration, temperature,
and dew point, and to cycle air through the chamber. The above-ground chamber is
attached to an insulated, water-tight, 6.4 mm thick aluminum soil lysimeter (1 m x 2 m x
1 m deep) which sits below ground level in a steel pit liner. The inside of the lysimeter
is painted with white nontoxic epoxy paint covered with clear, 2.5 mil oriented-strand
adhesive-backed Teflon™. Lysimeters are filled with a reconstituted (top, middle, and
lower horizons) coarse textured sandy loam soil topped with forest floor litter. Soil
temperature is measured with thermistors and soil moisture is measured with Time-
Domain Reflectometry probes (TDR, Campbell Scientific, Logan, UT). Outside
(chamberless "CL") plots have lysimeters but no enclosure.
Chamber environmental conditions, i.e., CO2, air temperature, and dew point, are
based on ambient data collected at a weather station near the chambers. Data are
transferred each minute to processors at individual chambers and used to determine target
levels inside the chambers. Targets can be either ambient or elevated above ambient, i.e.
+200 jil H CO2 and/or +4CC in the current experiment. After planting on 6 and 7 June,
1993, trees were exposed to ambient CO2 and temperature until late August, 1993 when
the CO2 and temperature treatments were imposed. Climate data for this paper is for
1995, based on 1-min. averages. Data are reported here for ambient CO2 and temperature
chambers (ACAT). Weather station data are used for CL plots.
2.2	Tree Seedling Selection and Culture
Seedlings (1 + 1's, i.e. two years of growth) were obtained from the Weyerhaeuser
Company in Aurora, OR, and grown from mixed seeds ("woods run") from five low
elevation (<500 m) seed zones in the southern Willamette Valley and western Cascade
Mountains. Fourteen trees were planted in each chamber or CL plot in three east-west
rows of five (outer), four (middle) and five (outer) trees. Outer row trees are 18.5 cm
from the north and south ends of the chamber and 19.5 cm from the east and west ends of
the chambers. Middle row trees are 39.5 cm from the west and east ends of the chamber.
Rows are 36.5 cm apart and trees within rows are 39.5 cm apart.
2.3	Tree Response Measurement and Statistical Analsysis
For ACAT chamber or CL plot stem diameter is measured on each tree with a
Mitoya Digimatic digital caliper at a reference point just above the soil surface and first
nodal swelling on the tree. Other parameters are measured on each tree with a ruler or
meter stick: height from the diameter mark to the highest point on the terminal shoot,
terminal shoot length from bud scar to longest point, length for one randomly selected
needle on each tree midway along the terminal shoot, and terminal bud length from
attachment to stem to tip. Leaf area index (LAI, or unit leaf area/unit ground surface
area) is based on a function: leaf area = 1.65 + 1.682 * X - .086 * X2, where X=stem
diameter? * height. Measurements are made weekly to monthly for spring-summer
months; and bimonthly for fall-winter months. Branches numbers are determined from a
subset of four trees. Needle areas and specific weights are based on pooled samples for
all trees in a ACAT chamber or CL plot. Needles from this sample are measured for
carbon and nitrogen concentrations [flame combustion analysis (Carlo-Erba)].
This analysis considers three replicate ACAT chambers and two CL plots (CL).
Data from all available sampled trees were used to obtain ACAT or CL means. Statistical
analysis on means is by an unpaired t-test (StatView®, Abacus Concepts).

-------
3. Results
3.1	Chamber Environment
Climate data for 1995 indicate that CO2, air temperature, and dew point are very
similar for the ACAT vs. CL trees. Annual averages for ACAT and CL trees,
respectively; were: CO2 of 410 and 394 jimol moH; air temperatures of 12.6 and 12.3°C;
and dew points of 6.9 and 7.4°C. Annual growing degree days (5°C base) were 2816
and 2750, respectively, for ACAT and CL trees. Soil temperature for ACAT trees closely
follows air temperature (data not discussed here). These results confirm the initial
performance characteristics of the chambers for Nov. 1993 through Nov. 1994 (Tingey et
al., 1996).
The difference in CO2 concentrations is due primarily to higher levels (= 50 (imol
moH) at night for ACAT compared to CL trees due to the difficulty in fully venting the
respiratory buildup in CO2 in the chambers. Air temperature levels differ especially on
cold winter nights when ACAT temperature cannot go much below 0°C due to the nature
of the propylene glycol solutions used for temperature control. Furthermore, ice may
form on chamber cooling coils when they approach 0°C, making them less efficient in
removing heat and resulting in an increase in air temperature above ambient.
The major differences in environment between ACAT and CL trees are soil
moisture, vapor pressure deficit (VPD), photosynthetically active radiation (PAR between
400-700 nm), and wind velocity. Soil moisture in chambers is dependent on irrigation
which is quantitatively applied to mimic a normal seasonal (wet winter and dry summer)
cycle for Pacific Northwest forests (Tingey et al.. 1996). The ACAT trees receive only
irrigation water, resulting in 821 mm of water applied in 1995 compared to a total of 2270
mm for CL trees which receive both irrigation and rainfall. The difference in water
received results in lower soil moisture levels during the early and late growing season for
ACAT compared to CL trees (Figure 1A), Based on the annual average air temperature
and dew point data, annual average calculated VPD was slightly larger for ACAT vs. CL
trees, at 0.47 and 0.40 kPa. The PAR varied diurnally and seasonally (Figure 2B, data
for second half of 1995) across chambers due to shading and/or reflectance from chamber
structural supports and absorption by the Teflon™ as well as condensation of moisture on
the chamber walls during cool nights. Average PAR was 12% lower for ACAT
compared to CL trees. Wind velocity for ACAT trees is at a constant rate and averaged
0.27 m s-> within and 0.61 m s-' above the tree canopy; resulting in = 10 air changes per
minute through the chambers (Tingey et al., 1996). In contrast, outside wind velocity
varies diurnally and seasonally, with a average speed of = 2.6 ± 2.7 m s-1 , based on
daily aveages for approximately May-September 1994 from the for Corvallis area.
3.2	Plant Response
Stem diameters were first noticed to be significantly smaller (p<0.05) for ACAT
compared to CL trees on 30 June 1994 (day 389, Figure 2A), becoming ~ 3 mm less for
ACAT than for CL trees by 18 October 1994 (day 489). This difference persisted
through 11 December 1995 (day 918), when there was a significantly smaller stem
diameter for ACAT compared to CL trees (Table 1). In contrast, heights were similar for
ACAT and CL trees until 15 June 1995 (day 739, Figure 2B). Thereafter heights tended
to be smaller for ACAT than for CL trees due to shorter main flush terminal shoots, but
the differences in height and terminal shoot length were not statistically significant on day
918 (Table 1). The LAI was significantly lower for ACAT than for CL trees (Table 1).
Branch numbers were significantly lower and terminal buds significantly shorter for
ACAT than for CL trees (Table 1). However, needle characteristics (lenght, area, specific
weight, % C and N) did not differ between ACAT and CL trees (Table 1).

-------
A. Soil H20
ACAT
/
ACAT Minus CL
B, PAR
o 1500
~ ACAT
q ACAT
Minus CL
* 11 j 11 * 11
300 350
150 200 250
Julian Date
250
Julian Date
Figure 1. Plant available soil water (panel A), and photosynthetically active
radiation (PAR) above the tree canopy (panel B), and in ambient chambers
(ACAT) and difference between ACAT and outside air (CL) in 1995. Data are
means for three ACAT chambers and two CL plots.
2
o
E
 to
o
o
o o
o o
00 03
O
o
o
Days From Planting
OS
0)
I
900-
800-
700-
600-
500-
400-
OOCO o
~ ACAT
C CL
O
o
/•
0 St •
*
*
B. Height
:s
300 I I'll [ ) I I ) [ H U | I I M | I I I I [ I I I I | I M I | 1 I I'I [ It I I [ M I »
OOQOOOOOOOO
OOQoOOOO
eOTWtof-eoojo
O o
o o
CM
Days From Planting
Figure 2. Stem diameter (panel A) and shoot height (panel B) for Douglas-fir
seedlings growing in ambient chambers (ACAT) or outside air (CL) from 7
June 1993 (day 0) through 11 December 1995 (day 918). Data are means for
three ACAT chambers and two CL plots. Average coefficients of variation for
replicates across dates for ACAT and CL, respectively, are 4.9 and 4.3% for
stem diameter and 3.6 and 3.3% for height.
4.0 Discussion
Over the first 2 1/2 years of the study, ACAT chamber trees grew well but were
slightly smaller than CL trees. The differences in general growth may be attributed to

-------
Table 1. Douglas-fir shoot responses in ambient CO2 and temperature chambers
(ACAT) and outside air (CL). Data from fall, 1995; dates between 862 to 918
days after planting depending on parameter. Values are means ± standard error
for 3 ACAT and for 2 CL. Parameters averaged for 14 plants per chamber or
plot; except for branch counts averaged for 4 plants per chamber or plot; and
needle area, specific weight, nitrogen, and carbon based on needles pooled from
all 14 plants, Terminal shoot and needle lengths are for main flush. The p
values are from unpaired t-tests.
Parameter
ACAT
CL
p Value
Stem diameter (mm)
18.7 ± 0.4
21.7 ± 0.2
0.016
Height (mm)
774 + 29
862 ± 32
0.143
LAI
3,03 ± 0.06
4.00 ±0.15
0.006
Terminal Shoot Length (mm)
248 ± 15
301+ 3
0.075
Terminal Bud Length (mm)
10 ± 0.3
11 ± 0.2
0.032
Branch Count (#)
175 ± 14
264 + 15
0.024
Needle Length (mm)
28 ± 1,3
26 + 0.4
0,205
Needle Area (cm2)
0.28 + 0.01
0.25 ± 0.01
0.253
Needle Specific Wt. (g cm-2)
0.012 10.003
0.013 ± 0.004
0.181
Needle Nitrogen Cone, (% dry wt.)
1.46 ± 0.05
1.48 ± 0.04
0.816
Needle Carbon Conc.(% dry wt).
47.6 ± 0.9
47.6 ± 0.5
0.992
higher moisture stress (lower soil moisture coupled with higher VPD) and lower solar
radiation for ACAT than for CL trees, both of which could reduce plant growth. The
TREEGRO (Weinstein and Beloin, 1990) model is being used with the climate data for
the chambers and outside plots to test this hypothesis. For woody plants, the small
differences in early growth, as found for Douglas-fir, will likely become greater over time
as shown for orange trees in open-top chambers (Olszyk et al,, 1992). The slightly
greater bud length at the end of 1995 suggested that the CL trees would start off with
higher potential productivity for 1996, and this has been occurring (data not shown here).
There were no obvious differences in needle characteristics which would explain
difference in growth for ACAT compared to CL trees. The same nitrogen concentrations
in ACAT and CL trees suggest that atmospheric nitrogen deposition, which is an
uncontrolled variable expected to occur only for CL and not for ACAT trees, is not a
determining factor for nitrogen availability and subsequent tree growth in this study.
Increased leaf nitrogen due to deposition could have affected overall tree growth due to
the relatively low initial fertility of this soil and reliance on soil processes to provide
nutrients in this study. Lack of differences in leaf carbon concentrations and
photosynthetic rates (data not shown here) for ACAT and CL trees imply that carbon
assimilation is not affected by the chambers. Essentially the same nitrogen concentrations
for ACAT and CL needles also suggest similar carbon assimilation as leaf nitrogen is

-------
closely correlated with photosynthetic rates (Field and Mooney, 1986). There is little
information on dark respiration for needles and none for branches in this study, thus,
effects of the chambers on overall carbon balance of the shoots could not be determined.
We recommend that the effect of the experimental system itself on plant response be
an important consideration whenever any highly environmentally controlled facility is
used, as discussed in detail for past studies using open-top chambers and ozone (Heagle
et al,, 1988). Our analysis to date shows that while some basic tree growth responses are
different in the chambers vs. outside air; these difference are likely attributable to
documented environmental conditions— e.g. soil moisture, VPD, and light. Further
analysis is necessary, but data to date show no evidence that basic tree responses to the
imposed treatment variables (C02 and temperature) are different in the chambers vs.
outside air, thus data obtained with the chambers can be used to determine effects on
Douglas-fir (or any other plant) to environmental stress, especially global change.
5.0 Acknowledgement
The information in this document has been funded wholly by the U.S.
Environmental Protection Agency. It has been subject to the agency's peer and
administrative review. It has been approved for publication as an EPA document.
Mention of trade names or commercial products does not constitute endorsement or
recommendation for use. Soil moisture data from Dr. Mark Johnson, prepared under
Dynamac Contract # 68-C6-0005 with the U.S. EPA, are greatly appreciated.
6.0 References
Allen, H.A., Jr., Drake, B.G., H.H. Rogers and J.H. Shinn. 1992. Field techniques
for exposure of plants and ecosystems to elevated CO-) and other trace gases. Critic.
Rev. Plant Sci. 11:85-119.
Field, C, and H.A. Mooney, 1986. The photosynthesis-nitrogen relationship in wild
plants. Pg. 25-56, In: Givnish T.J. (ed.) On the Economy of Plant Form and
Function. NY: Cambridge University Press.
Heagle, A.S., L.W. Kress, P.J. Temple, R.J. Kohut, J.E. Miller and H.E. Heggestad.
1988. Factors influencing ozone dose-yield response relationships in open-top field
chamber studies, pp. 141-179. In: Heck, W.W., O.C. Taylor and D.T.Tingey,
Eds. Assessment of Crop Loss From Air Pollutants. Elsevier, New York.
Jones, P., J.W. Jones, L.H. Allen, Jr. and J.W. Mishoe. 1984, Dynamic computer
control of closed environmental plant growth chambers: design and verification.
Trans. ASAE 27:879-888.
Olszyk, D M.. B.K. Takemoto, G. Kats, P.J. Dawson, C.L. Morrison, J. Wolf Preston
and C.R. Thompson. 1992. Effects of open-top chambers on 'Valencia' orange
trees. J. Environ. Qual. 21:128-134.
Taylor, G.E. Jr. 1994. Controlled environment facilities in ecology and environmental
sciences. Bull. Ecol. Soc. Am. 75:277-280.
Tingey, D.T., B. McVeety, R. Waschmann, M. Johnson, D. Phillips, P.T. Rygiewicz
and D.Olszyk. 1996. A versatile sun-lit controlled-environment facility for studying
plant and soil processes. J. Environ. Qual. 25:614-625.
Weinstein, D.A. and R. Beloin. 1990. Evaluating effects of pollutants on integrated tree
responses: A model of carbon, water and nutrient balances. Pp. 313-323. In.:
R.K. Dixon, R.S. Meldahl, G.A. Ruark and W.G, Warren (eds.), Process
Modeling of Forest Growth Responses to Environmental Stress. Timber Press,
PortlandrOR.

-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverts before compl
1. REPORT MO. 2
IPA/600/A-96/077
*
4. title and subtitle
Environmental modification and shoot growth in
a closed ecosystem to evaluate long-term
responses of tree seedlings t-n st-rpsn
5 REPORT OATE
6. PERFORMING ORGANIZATION COOE
7.AUTHORIS)
D.M. Olszyk, D.T. Tingey
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME ANO ADDRESS
USEPA, NHEERL, Corvallis, OR,
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME ANO AOORESS
US EPA ENVIRONMENTAL RESEARCH LABORATORY
200 SW 35tti Street
Corvaitis, OR 97333
13. TYPE OF REPORT AND PERIOD COVERED
Oifmnnoiiim P "ITl (*>"""
14. SWWSORTOS <5G E N$ftoDE
IS. SUPPLEMENTARY NOTES
1996. International Symposium on Plant Production in Closed Ecosystems, Aug 26-29,
1996, Narita, Japan
16. ABSTRACT
Determining ecosystem responses to stress requires careful control of
the environment while measuring biological effects. Technologically
advanced sun-lit closed chambers at WED were found to provide
appropriate environmental control and measurement compared to
chamberless controls. Chamber impacts on the ecosystems were
documented so that treatment effects could be interpreted. Differences
in growth can likely be attributed to less soil moisture and lower
light intensities, and possibly slightly greater vapor pressure
deficits in chambers vs. outside during parts of the year. Because of
the advance control system, the effects of these sunlit chambers on
plant growth appears to be negligible.
17. KEY WORDS ANO DOCUMENT ANALYSIS
a. DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
e. COSATi Field/Group
Controlled environment, vegetation,
Douglas-fir, Pseudotsuga menziesii


18. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report)
21. NO. OF PAGES
6
20. SECURITY CLASS (Thispage)
22. PRICE
EPA Perm 2220—1 (R«». 4-77) previous edition 
-------