young potted holm oak (Quercus ilex L.) trees. Received: 20 November, 1995 / Accepted: l0 Mm'ch, 1996. Abstract We investigated the effects of a short period ...
Oecologia (1996) 107:456-462
9 Springer-Verlag 1996
N. B e r t i n 9 M . S t a u d t
Effect of water stress on monoterpene emissions from young potted holm oak (Quercus ilex L.) trees
Received: 20 November, 1995 / Accepted: l0 Mm'ch, 1996
We investigated the effects of a short period of water stress on monoterpene emissions from Quercus ilex, a common oak species of the Mediterranean vegetation and a strong emitter of monoterpenes. The experiment was carried out on two young saplings with a branch enclosure system under semi-controlled conditions. Under unstressed conditions, small qualitative (cis[3-ocimene, trans-~3-ocimene, [3-caryophyllene and 1,8cineol) and large quantitative (as much as 40% for the main compounds emitted) differences were observed between the two apparently similar trees. Nevertheless these differences did not affect the short- and long-term responses to temperature and water stress. Daily courses of emissions and gas exchanges were similar before and after the stress. During the most severe stress, emissions were reduced by a factor of two orders of magnitude and the log-linear relationship between emissions and temperature no longer existed. Photosynthesis and transpiration rates decreased as soon as the soil started to dry, whereas monoterpene emissions slightly increased for few days and then dropped when the daily CO 2 balance approached zero. We concluded that under water stress monoterpene emissions were highly limited by monoterpene synthesis resulting from a lack of carbon substrate and/or ATR After rewatering, both emissions and gas exchanges recovered immediately, but to a level lower than the pre-stress level. These results have many implications for monoterpene emission modelling in the Mediterranean area, since the dry period generally extends from May to August. If our results are confirmed by field experiments, water stress could lead to a large overestimation of the emissions under summer conditions, when the algorithms based on light and temperature would give high emission rates. Abstract
N. Bertin 9M. Staudt (~)1 Host Institute: Joint Research Centre, Environment Institute, TP051, 21020 Ispra (VA), Italy Present address: I Universit~tHohenheim, Institut ftir Botanik, D-70593 Stuttgart, Germany
Key words Quercus flex 9 Holm oak - Monoterpene emission 9Water stress 9Mediterranean vegetation
Introduction Monoterpenes are biogenic substances with a common carbon skeleton (C10), derived from the same branch of isoprenoid metabolism (Croteau 1987). Besides their ecological functions in defense against pathogens and herbivores (Gershenzon and Croteau 1991), pollinator attraction (Knudsen and Tollsten 1993) and intra- or interspecies allelopathy (Tarayre et al. 1995), they are relevant in tropospheric chemistry. Like isoprene (hemiterpene), monoterpenes that are naturally emitted by vegetation are highly reactive and are thereby involved in processes such as aerosol and ozone formation, changes in CO and OH concentrations and acid deposition (Fehsenfeld et al. 1992). The numerous inventories that attempt to quantify the total annual emissions on a regional or global scale are based on land use area, leaf biomass factor, emission factors and environmental factors (Zimmerman 1979; Lamb et al. 1987, 1993; Pierce et al. 1990; Guenther et al. 1995). The annual biogenic hydrocarbon emission estimates are generally inaccurate by a factor of about 3 (Lamb et al. 1987; Pierce and Waldruff 1991). For monoterpenes, this uncertainty is mainly attributed to the poor knowledge of the physiological and environmental controls of plant emissions and to the high spatial and temporal variability (Fehsenfeld et al. 1992). During the last few years, new species and new areas have been investigated in many European ecosystems: e.g. the BEMA project in the Mediterranean area (BEMA 1994). They should help to quantify the annual emissions at the regional scale and to check the validity of algorithms developed from laboratory experiments. Whereas isoprene emissions clearly depend on both light and temperature (Tingey et al. 1981; Monson and Fall 1989), the role of other factors than temperature in the control of monoterpene emissions is still under discussion. Whatever species is investigated, monoterpene emissions increase exponentially with leaf temperature
O E C O L O G I A 107 (i996) 9 Springer-Verlag
(e.g. Dement et al. 1975; Tingey et al. 1980; Lamb et al. 1985). Nevertheless many investigations of diverse species have suggested the contribution of other controlling factors: relative air humidity (Dement et al. 1975), foliar moisture (Lamb et al. 1985), leaf development (Guenther et al. 1991), leaf oil content (Lerdau et al. 1994), genetic variability (Hanover 1966), season (Pier 1995; Staudt et al. 1995a), diurnal rhythms (Yokouchi and Ambe 1984) or light intensity (Steinbrecher 1989). Recently, the role of light in the short-term control of monoterpene emissions has been clearly established for Quercus ilex, which is a strong emitter of monoterpenes despite the absence of storage organs (Staudt et al. 1993). Field (BEMA 1994) and laboratory experiments on this widespread Mediterranean oak species (Bacilieri et al. 1993) later confirmed these facts at branch (Staudt and Seufert 1995) and leaf levels (Loreto et al. 1996a). Recent experiments (Loreto et al. 1996b) provided evidence that monoterpenes emitted by Quercus ilex leaves are formed from photosynthesis intermediates. In the continuity of the BEMA project, we investigated the behaviour of Q. ilex under water stress conditions. Indeed, soil water availability is one of the prevailing environmental factors under Mediterranean summer conditions (BEMA 1994), when high temperature and light intensity favor plant emissions. Therefore it should be considered in any attempt to model the annual emissions of Mediterranean vegetation. A dynamic branch enclosure system was used to measure monoterpene emissions, CO 2 assimilation and transpiration of young holm oak trees undergoing water stress for 18 days. This allowed us to characterize the long-term effects of drought stress on monoterpene emissions as well as the diurnal patterns during soil drying, and gave rise to a new view of the control of monoterpene emissions and the prevalence of temperature algorithms in the inventories. Materials and methods Plant material and biomass The experiment was carried out at the Joint Research Centre (Italy) on two 10-year-old holm oak saplings originating from Tuscany (Italy), which will be referred to as Q1 and Q2. An 18-day water stress period was applied alternately to the two trees. One tree (first Q2) was water-stressed while the other one was well-watered and vice versa. Trees were cultivated in 25-1 pots, placed in a greenhouse 4 weeks before the beginning of the experiment. On 20 December the top of each tree was enclosed in two 20-1 dynamic gas exchange chambers described in the next section). The
Table 1 Leaf age class distribution of leaf area and dry weight at the end of the experiment
457
warm and well-lit environment induced the onset of leaf development. Withholding water from Q1 at the end of January stopped leaf development until end of March. In contrast, Q2, which was water-stressed before Q1, started shooting in mid-February, about 1 week after rewatering. At the end of the experiment leaf area was measured with an optical planimeter (Delta-T device) and wood and leaf dry weights were determined after 10 days at 80~ in a ventilated oven. The progressive increase in leaf area was estimated a posteriori, according to observations taken throughout the experiment and to the distribution of area and weight in each leaf age class, including some old leaves that had fallen in the chamber and new leaves that appeared during the experiment. At the end of the experiment, new leaves represented 6 and 17% of the total enclosed leaf area for Q1 and Q2 respectively. These changes in the total leaf area were used to calculate the emission and gas exchange rates. At the end of March, the leaf age-class distribution of both branches indicated many small 1-year-old leaves and generally smaller and thicker leaves (higher specific leaf weight) on Q1 compared to Q2 (Table 1).
Branch exposure system The branch exposure system consists of two dynamic mass balance gas exchange chambers and a gas and climate monitoring unit. Each chamber consists of a cylindrical plexiglass frame (40 cm long x 25 cm diameter) sustaining a thin Teflon bag (50 gm thick). A controlled flow (pump and mass flow controllers, MKS, Germany) of charcoal-filtered air (ozone and monoterpene free) enters in the upper part and exits through the stem insertion port at the base of the chamber. Air flow was set to 15-20 1 min 1 (maximum 1 min air residence time) according to the external temperature. A stainless steel electric fan placed at the top of each chamber ensured the homogenization of the internal atmosphere. Two highpressure sodium lamps installed above each chamber provided photosynthetically active radiation (PAR) of t000 gmol m -2 s 1 from 9 a.m. to 5 p.m, measured by three PAR quantum sensors (PAR-SB 190, licor) positioned halfway up outside of the chambers. Chamber temperature was indirectly controlled by air conditioning of the whole greenhouse. Air and soil temperatures were measured by PT1000 sensors. In each pot a tensiometer, installed at a depth of about 20 cm, indicated the soil water availability. These sensors were limited to -80 kPa and therefore the absolute values during advanced water stress were unknown. CO 2 and H20 concentrations of inlet and outlet air were determined by an absolute CO 2 analyzer (BINOS 100, Leyboldt) and a dew point mirror (MTS-2, Walz). Climatic and gas data were scanned every 5 and 1 s and recorded every 15 and 1 min respectively on an acquisition unit (Logger Delta-T). Finally, CO 2 assimilation, transpiration and stomatal conductance were calculated according to von Caemmerer and Farquhar (1981). Monoterpene emission rate was calculated as the difference between outlet and inlet air concentrations multiplied by chamber flow and divided by the projected leaf area. Monoterpene sampling and analysis Monoterpenes were trapped in glass tubes (Chrompack, 15 cm long and 3 mm inner diameter) filled with approximately 125 mg
Leaf number
Leaf area cm 2
Leaf dry weight g
Specific leaf weight g cm -2
Q1 Leaves 93 Leaves 94 Leaves 95
25 48 9
138.8 270.9 24.3
2.85 4.30 0.44
0.0206 0.0159 0.0182
Q2 Leaves 93 Leaves 94 Leaves 95
26 22 13
146.3 196.1 73.0
2.68 2.84 0.97
0.0183 0.0145 0.0132
458
OECOLOGIA 107 (1996) 9 Springer-Verlag
Tenax TA (Aldrich, 20-35 mesh) and placed in a cooled sampling device (Staudt et al. 1995b). Defined volumes of air were sampled at the inlet and outlet ports of the chambers, by means of a pump and two mass flow controllers (Brooks, 0-0.5 1 min -1) connected to a timer (3 1 air at a rate of 200 ml min -1 for our experiment). A prepurging time of 5-10 min on a bypass line ensured the sample quality. Air samples were analysed by a gas chromatograph (GC CP9001, Chrompack) fitted out with a desorption unit (TCT/PTI CP4001, Chrompack), a fused silica capillary column (25 m x 0.32 mm, df: 1.2 p,m CP-Sil 8 CB, Chrompack) and a flame ionization detector. The carrier gas was helium (85 kPa) and the temperature program presented the following sequence: 3 rain precooling at -100~ 10 rain desorption at 200~ 1 min injection at 200~ GC-oven: 4 rain at 65~ 2.5~ min -1 to 80~ 2.0~ min -1 to 100~ and 20~ min -1 to 240~ Gaseous and liquid calibration standards prepared from commercial authentic monoterpenes of high purity (Fluka Aldrich, 95-99% purity) allowed the peak identification and quantification. Sampling and analysis procedures have been described in details by Staudt et al. (1995b).
25
. 1,2
20.
0,8
~,
15.
0,4
"2
10.
g,
0,0
300
B 225
'~ = '~
150
75
0
-~'
9.
Experimental design and protocol Two weeks after installation of the chambers, water stress lasting about 18 days followed by rapid rehydration was successively applied to both trees. Water was withheld from Q2 from 5 to 24 January and from Q1 from 24 January to 12 February. On the last day of water stress the tree was copiously irrigated and then regularly watered on the following days. Soil drying up was somewhat faster for Q1. Throughout the experiment air and soil temperatures ranged from 17 to 30~ and from 10 to 22~ respectively, and the differences between the two chambers were less than l~ Throughout this period, inlet and outlet air of each chamber was sampled every 2 or 3 days, three times a day between 10 a.m. and 4 p.m. at a constant temperature (25~ _+ 0.2) and PAR (1200 + 100 gmol m -2 s-l). The chamber temperature was maintained at 25~ for 1 h before and during monoterpene sampling. Leaf temperature was measured only occasionally during the illuminated period with two thermocouples (Ni/CrNi type). The values showed a slight but stable exess over air temperature of 0.5-1.5~ On five particular days, before (5 January), during (24 January, 1 and 12 February) and after (28 February) each water stress period, we measured monoterpene emissions every 60-90 min from 8 a.m. to 6 p.m. and three times in the night without controlling the temperature, to characterize the natural diurnal profiles at different stages of soil drying.
Results General pattern of monoterpene emissions and gas
~,
-T
-
9
of~,,
.
6.
1,8
3
