Irradiance and temperature affect the competitive interference of blackberry on the physiology of European beech seedlings

Authors

  • Mariangela N. Fotelli,

    1. Institute of Forest Botany and Tree Physiology, Albert Ludwig University of Freiburg, Georges-Köhler-Allee, Gebäude 053/054, 79110 Freiburg i. Br., Germany
    Search for more papers by this author
  • Peter Rudolph,

    1. Institute of Forest Botany and Tree Physiology, Albert Ludwig University of Freiburg, Georges-Köhler-Allee, Gebäude 053/054, 79110 Freiburg i. Br., Germany
    Search for more papers by this author
  • Heinz Rennenberg,

    1. Institute of Forest Botany and Tree Physiology, Albert Ludwig University of Freiburg, Georges-Köhler-Allee, Gebäude 053/054, 79110 Freiburg i. Br., Germany
    Search for more papers by this author
  • Arthur Geßler

    Corresponding author
    1. Institute of Forest Botany and Tree Physiology, Albert Ludwig University of Freiburg, Georges-Köhler-Allee, Gebäude 053/054, 79110 Freiburg i. Br., Germany
      Author for correspondence: Arthur Geßler Tel: +49 761, 2038309 Fax: +49 761, 2038302 Email: arthur.gessler@sonne.uni-freiburg.de
    Search for more papers by this author

Author for correspondence: Arthur Geßler Tel: +49 761, 2038309 Fax: +49 761, 2038302 Email: arthur.gessler@sonne.uni-freiburg.de

Summary

  • • The potential negative influence of competition from early successional species like blackberry (Rubus fruticosus) may be decisive for the natural regeneration success of drought-sensitive beech (Fagus sylvatica), especially in the light of climate change.
  • • With a split plot glasshouse experiment, we investigated the influence of two air temperature and irradiance levels on the competitive interference of blackberry on the water, nitrogen (N) and carbon (C) balance of beech seedlings under moderate drought.
  • • When increased temperature was accompanied by low irradiance the biomass, root-to-shoot ratio, N uptake and assimilation rates of blackberry were lower compared with beech, either grown alone or with blackberry. By contrast, when elevated temperature and high irradiance were combined, the root-to-shoot ratio and specific N uptake rate of blackberry were substantially increased, while the N acquisition of beech was impaired. Under lower temperature, with either full light or shade, the presence of blackberry had no significant effects on beech, for almost all tested parameters.
  • • Under elevated air temperature beech was impaired by the presence of blackberry at high irradiance. These findings emphasize the interacting effects between environmental factors and competition on the establishment of beech regeneration, which should be considered for future forest management in the frame of climate change.

Introduction

European beech is the most abundant forest tree of the potential natural vegetation in Central Europe (Ellenberg, 1992). It is strongly promoted by current forest policy (Tarp et al., 2000) in order to support the development of stable, mixed conifer-broadleaf forests, contrary to the previously dominating coniferous monocultures. Forest management practises, such as thinning by means of selective felling of trees, are applied in Central and Northern Europe for enhancing the natural regeneration of beech (Dertz, 1996, Ministerium für Ländlichen Raum, Ernährung Landwirschaft und Forsten in Baden-Württenberg, 1999; Tarp et al., 2000), by increasing nutrient, water and light availability in the forest understory (Breda et al., 1995; Aussenac, 2000; Mizunaga, 2000; Thibodeau et al., 2000; Aranda et al., 2004).

However, beech is a drought-susceptible species (Ellenberg, 1992; Peuke et al., 2002) and the predicted global warming combined with increased, in frequency and duration, drought periods during the growing season (IPCC, 2001) could impair its performance (Geßler et al., 2001; Fotelli et al., 2002a; Fotelli et al., 2003). Forest canopy opening and, thus, increase of irradiance, is considered to positively affect the growth of beech seedlings after their first establishment (Welander & Ottosson, 1998; Collet et al., 2001; Aranda et al., 2004). However, this practise can also have negative effects on the water balance of beech at warm and dry sites (Geßler et al., 2004). In addition, elevated irradiance as a result of thinning could influence carbon assimilation of different understorey species in a different way, hence altering intraspecific competition.

Therefore, another factor with a potential negative impact on the success of beech natural regeneration could be the increasing density of understory vegetation after opening of the forest canopy (Fotelli, 2002). In particular, early successional, fast growing species with high competitive ability could induce unfavourable conditions during the early growth phases of beech regeneration, especially when soil water availability is low. Previous studies, assessing the effect of competitive interference by the early successional, rather invasive, blackberry on young beech seedlings, showed that both their water and nitrogen status were seriously impaired by the presence of blackberry when water availability was limited (Fotelli et al., 2001; Fotelli et al., 2002b).

Given that climate change scenarios foresee elevated air temperatures in the near future (IPCC, 2001) having major effects on the physiology of trees (BassiriRad, 2000; Saxe et al., 2001), air temperature is an additional important parameter that may influence the competitive interference between early successional, invasive species and young beech seedlings. Fotelli et al. (2004) hypothesized that silvicultural treatments that stimulate growth and improve the nutrient and water balance of beech seedlings at present might have adverse effects on the physiology of natural beech regeneration under the warmer and drier climate conditions expected in future. However, under field conditions no clear distinction can be made between the direct effects of abiotic factors and effects of alterations in competition patterns between beech and neighboring species.

In order to contribute to an integrated characterisation of such changes in competition patterns under changing environmental conditions, the following experimental setup was chosen for this study: According to previous studies (Fotelli et al., 2001, 2002a) we simulated moderate drought and additionally applied two different levels of air temperature and irradiance, and two different regimes of competitive interaction (beech without competitor, beech plus blackberry). Under these conditions we assessed the water, nitrogen and carbon balance of beech. Our working hypothesis was that the presence of a fast-growing competitor such as blackberry adversely affects the water and N balance of beech under restricted water availability as applied in the present study. Based on previous field studies (Fotelli et al., 2004) beech was expected to be less competitive at elevated temperature and irradiance.

Materials and Methods

Experimental design

Young European beech (Fagus sylvatica L.) seedlings of similar height (c. 0.30.4 m) and structural characteristics, produced in a forest nursery in Freiburg, Germany (longitude: 48° N, latitude: 7°51′ E, 275 m asl) and blackberry (Rubus fruticosus) plants, collected in winter 2000 from the field site ‘Tuttlingen’, located in southern Germany in a low mountain range (Schwäbische Alb, Germany; longitude: 8°50′ E, latitude: 47°58′ N, 740–760 m asl; for a detailed site description see Keitel et al., 2003) were used for the experiment conducted in a glasshouse. One-year-old beech seedlings were chosen in order to examine the early phases of seedling establishment being most critical for survival and development of beech in the forest understorey (Madsen & Larsen, 1997). Parts of blackberry shoots (c. 0.05–0.1 m height) were collected in winter 2000 and were transferred from the field site to the glasshouse where they were left in tap water to develop new roots. In winter 2001 blackberry and beech seedlings were transplanted into experimental containers (diameter 0.45 m; depth 0.35 m) simultaneously and were grown thereafter under controlled conditions (air temperature, air relative humidity, irradiance). The time period of 1 yr between the collection of blackberry shoots from the field site and the initiation of the experiment was considered adequate for ensuring acclimation of the evergreen blackberry to the new light conditions of the glasshouse.

The planting substrate was a homogenous mixture consisting of soil collected from the upper 0.02 m of the profile at the field site and quartz sand (1 : 1). By mixing soil with sand uncontrolled nitrogen release due to mineralization under glasshouse conditions was minimised and the field pH (5.7–7.5) was maintained (Fotelli et al., 2002b). The soil of the field site is characterised as Terra fusca – Renzina derived from limestone (Weißjura beta and gamma series). The mixture of top soil and quartz sand had a water holding capacity of c. 20%. The soil water potential in all experimental basins was steadily measured with gypsum blocks (5201F1 G-Blocks and 5910-A Soil Moisture Meter; Soil Moisture Equipment Corp., Santa Barbara, CA, USA). By regulating irrigation accordingly, soil water potential was kept within a range of −0.2 to −0.5 MPa (Fig. 1b), conditions corresponding to the soil water availability found in the field site during relatively dry periods (Geßler et al., 2001). This approach has also been used in similar previous glasshouse experiments (Fotelli et al., 2001; Fotelli et al., 2002b).

Figure 1.

Patterns of soil temperature (a) and soil water potential (b) during the experiment. Two temperature regimes were applied (C, cold; W, warm) with a temperature difference of 1–6°C between them. Additionally, two irradiance levels were applied within each temperature regime (L, full light, where the sum of day light plus light supplement was provided to the plants; S, shade, where mean irradiance was 50% of the full light level). The combination of temperature and irradiance regimes resulted in four different growth regimes (CS, filled squares; CL, filled circles; WS, open squares; and WL, open circles). A controlled irrigation regime, initiated on 12 February was applied in order to maintain soil water potential within a range of –0.2 to –0.5 MPa, which corresponds to the intermediate irrigation regime applied in a previous study (Fotelli et al., 2001). 15N tracer was applied at the initiation of the irrigation regime. Each value represents the mean temperature and soil water potential of six independent basins, where the plants were growing.

A split-split-plot design was applied with the following components: air temperature represented the main plot, while irradiance was a split-plot within air temperature and competition was a split-plot within the growth regimes produced by the combination of air temperature and irradiance.

The air temperature level was defined by the application of either high air temperature (warm-W) or lower air temperature (cold-C), with a steady difference of c. 6°C between them. Two separate growth rooms of the glasshouse were used for the application of the temperature levels; each room represented one temperature regime. Air temperature manipulation was enabled by the central unit of the glasshouse, providing full control on environmental conditions. Moreover, the air temperature level in the growth rooms was also steadily checked with a portable Thermohydrograph established in the center of each one.

The irradiance level was defined by the application of either full light (L), at which photosynthetically active radiation (PAR) fluctuated between 200 and 280 µmol m−2 s−1 from the bottom to the top of the plants, or shade (S), at which irradiance was decreased by c. 50%, compared with full light. Plants of the L level were grown under natural daylight of the glasshouse, supplemented with artificial light provided by mercury lamps (SON-T AGRO 400, PHILIPS GmbH) on a 16 h light-/8 h dark-phase rotation basis. The same light sources were used for the S level, but the PAR received by the plants was c. 50% reduced, compared with L level, by putting a shading net between plants and mercury lamps. In each of the two glasshouse rooms (temperature plots), the irradiance levels were applied by covering half of the plants with the shading net. The applied levels of irradiance were regularly controlled using the PAR – sensor of a steady state Li −1600 porometer (LiCor Inc., Lincoln, NB, USA) and were within the above-mentioned ranges in both glasshouse rooms throughout the experiment.

Kreuzwieser et al. (1997) showed that photosynthetic CO2 exchange of beech seedlings was light-saturated at a c. 200–250 µmol m−2 s−1 irradiance. Moreover, in thinned beech stands of the field site ‘Tuttlingen’, where invasion of the fast-growing blackberry is observed, irradiance varied between 50 and 350 µmol m−2 s−1 during the 2000 growing season (Fotelli et al., 2003). At the end of the season blackberry plants were collected for the present study. Hence, irradiance of the L light regime in the glasshouse provided light saturating conditions and was within the levels at which both, beech and blackberry, grow under natural field conditions.

The combination of the above-described air temperature and irradiance levels produced four growth regimes, characterised by the following environmental conditions during the experiment: first CL with air temperature during daytime between 13 and 20°C and irradiance between 200 µmol m−2 s−1 (plants bottom) and 280 µmol m−2 s−1 (plants top); second CS with air temperature during daytime between 13 and 20°C and irradiance between 100 µmol m−2 s−1 (plants bottom) and 140 µmol m−2 s−1 (plants top); third WL with air temperature during daytime between 20 and 26°C and irradiance between 200 µmol m−2 s−1 (plants bottom) and 280 µmol m−2 s−1 (plants top); and fourth WS with air temperature during daytime between 20 and 26°C and irradiance between 100 µmol m−2 s−1 (plants bottom) and 140 µmol m−2 s−1 (plants top).

Within each of the four growth regimes, two competitive interference levels were applied: in the first one (B), six beech seedlings were planted in a container, c. 0.12 m apart from each other. In the second one (B + R), six beech seedlings and six blackberry plants were planted together in a container (c. 0.08 m distance between blackberry and beech and c. 0.12 m distance between plants of the same species), in order to assess the effects of blackberry on the physiology of beech seedlings due to its potential competitive advantage and the space and resources limitation its presence induces. Three replicate containers were used for each competitive interference treatment and combination of temperature-irradiance regimes, building a total of 24 containers. From each container two plants per species were randomly selected among the six and all measurements and laboratory analysis were conducted thereafter on them.

Soil temperature was steadily measured with temperature sensors installed in each of the experimental containers (Pt 100 – Type GES 150; Greisinger Electronic GmbH, Regenstauf, Germany) and read with a GMH 2000 measuring instrument (Greisinger Electronic GmbH, Regenstauf, Germany). Fig. 1(a) shows that the different levels of air temperature produced clear differences in soil temperature between the C and the W level. Soil temperature of the C level was within the range observed in natural beech stands with closed canopy (Geßler et al., 1998, 2002). However, Fig. 1(a) also demonstrates that soil temperature was only slightly altered by shade within each main (air temperature) plot, indicating that environmental conditions, other than temperature, were similar between the main plots.

The experiment was initiated c. 7 wk after transplantation and c. 1–2 wk before bud break. After this period used for ensuring recovery from transplanting shock, the experiment was performed for 10 wk. An even shorter duration was found adequate for assessing abiotic effects on the competitive interaction between beech seedlings and blackberry in previous experiments (Fotelli et al., 2001; Fotelli et al., 2002b).

Growth parameters

At the end of the experiment the two plants per container selected for measurements were harvested and separated into stems, leaves and roots. Roots were washed twice for one minute with tap water to remove adhering 15N nitrate (see section 15N tracing below). The weight of the plant parts was measured immediately after harvest (fresh weight – f. wt) and after oven-drying for 3 d at 65°C (dry weight – d. wt). Dry weights of the plant parts were used to calculate root-to-shoot ratios. Moreover, the specific leaf area (SLA) was determined (m2 g−1 d. wt) in five representative leaves per plant.

Foliar biomass increment (d. wt) was defined by the difference in leaves dry weight between the end and the start of the experiment. The foliar dry weight of blackberry plants at the beginning of the experiment was determined by multiplying the number of leaves with the average dry weight of leaves, also used for the determination of SLA. It should be noted that beech seedlings had no leaves at the beginning of the experiment, since bud break was initiated shortly after.

15N tracing

Although preferential uptake of ammonium by roots of beech and other woody plants is reported (Plassard et al., 1991; Kronzucker et al., 1996; Geßler et al., 1998), nitrate is the main pedospheric N source for beech growing on limestone-derived soils with high pH (Fotelli et al., 2002a; Fotelli et al., 2004). The high pH values found in the soil of the field site ‘Tuttlingen’ used in the experiments, favours conversion of ammonium to nitrate by autotrophic nitrification. Thus, only very low ammonium concentration can be detected in soil extracts, whereas ammonium is generally not present in soil water (Geßler et al., 2004). Therefore, we assessed the patterns of nitrate uptake and partitioning in the studied plants by applying a NH415NO3 (98%15N enrichment) solution to the soil of the basins. The solution was applied uniformly on the surface of the soil mixture of all basins on 12 February 2001. In order to minimise the possibility of increasing plant available nitrogen in the soil and to avoid fertilisation effects due to the application of the 15N tracer, c. 41.80 mg NH415NO3 were used per m2 soil surface corresponding to < 3% of the natural inorganic N contents (cp. Fotelli et al., 2002b; Fotelli et al., 2004).

Plant 15N uptake, total N and δ13C isotope composition

After the plant material was harvested it was oven dried for 3 d at 65°C and ground with a ball mill into a fine homogenous powder. Between 1 and 2 mg aliquots of the different tissues (leaves, stems, roots) was transferred into tin capsules (Type A; Thermo Quest, Egelsbach, Germany) and injected into an elemental analyser (NA 2500; CE Instruments, Milan, Italy) coupled by a Conflo II interface to an isotope ratio mass spectrometer (Delta Plus; Finnigan MAT GmbH, Bremen, Germany). The total N content (%) is provided by the elemental analyser. The specific 15N uptake (µmol d−1 g−1 d. wt) was estimated using the following equation:

image

where 15Nt, 15Nc are the atom percentage concentrations of 15N in plants treated with the 15N-enriched solution and in control plants, respectively. Additional beech and blackberry plants not treated with 15N-addition were used as controls. [N] is the total N percentage concentration (of a g dry weight), MW is the molecular weight of 15N (g mol−1), and t is the time from 15N application until plant harvest (70 d). The total 15N incorporation in leaves, stems and roots was calculated by multiplying the specific 15N incorporation in these tissues with their respective dry biomass. For determining 15N incorporation per plant the 15N incorporation of the plant parts were summed up.

The δ13C values were defined as:

δ13C(‰) = [(Rsample/RPDB) − 1] × 1000

where Rsample and RPDB are the 13C : 12C ratios of sample and Pee Dee belemnite (PDB; Craig, 1957), respectively.

Transpiration and assimilation rates

Transpiration and assimilation rates were measured under the environmental conditions applied within each growth regime on three fully expanded leaves of the selected plants. Measurements were conducted between 08 : 00 h and 11 : 00 h on two consecutive days at the end of the experiment with a portable gas exchange measuring system (LCA-4, ADC, Hoddesdon, UK). In order to check the zero-point-setting between measurements, the machine performs an internal calibration for which air passes through columns containing Soda lime (ADC, Hoddesdon, UK) and Drierite (Hammond Corp., Xenia, OH, USA) to absorb CO2 and H2O, respectively.

Statistical analysis

All statistical analysis was carried out using SPSS 8.0 (SPSS, Inc., Chicago, IL, USA). Means were built from the two plants per species used for measurements from each container. The containers were then used as replicates for statistical analysis (n = 3 for each combination of growth regimes).

For each competitive interference treatment (B, B + R, R), the effect of temperature and irradiance was assessed using a General Lineal Model – Univariate Anova process. Moreover, within each growth regime (CS, CL, WS, WL), significant differences between the interference treatments were detected with one-way anova by using the Tukey HSD posthoc test.

Results

Growth parameters

Only irradiance had a significant effect on the foliar dry weight increment of the plants during the experiment (Fig. 2a). Within all growth regimes, R. fruticosus presented significantly lower foliar dry weight increment than beech seedlings. However, this difference became substantially smaller in the WL regime; then beech seedlings either with or without blackberry had limited foliar biomass increment, compared with the other growth regimes.

Figure 2.

Effects of irradiance and temperature (CS, cold-shade; CL, cold-full light; WS, warm-shade; WL, warm-full light) and competitive interference (B (Fagus sylvatica), beech as control; B + R, beech with Rubus fruticosus; R, R. fruticosus with beech) on foliar dry weight increment (a), total plant dry biomass (b), specific leaf area (c) and plant root-to-shoot ratio (d) of beech seedlings and R. fruticosus plants. Vertical bars indicate ± SE of the mean. Within each treatment, means of the competitive interference treatments are significantly different at a 95% level of significance, when they share no common letter. The effects of temperature (T) and irradiance (L) on each competitive interference treatment (B, B + R, R) were the following: (a) Foliar dry weight increment, B: T ns, L P = 0.034, TxL ns; B ± R: T ns, L ns, TxL ns; R: T ns, L ns, TxL P = 0.013 (b) Total dry biomass, B: T ns, L ns, TxL ns; B ± R: T ns, L ns, TxL ns; R: T ns, L ns, TxL P = 0.011 (c) Specific leaf area, B: T ns, L ns, TxL ns; B ± R: T ns, L P = 0.048, TxL P = 0.041; R: P = 0.003, L P = 0.001 TxL P < 0.001 (d) Root-to-shoot ratio, B: T ns, L(T) ns; B ± R: T P = 0.018, L P = 0.008, TxL P = 0.009; R: T ns, L ns, TxL P = 0.016 (ns: P > 0.05).

No significant effects of either air temperature, or irradiance were found on total dry biomass of any of the interference treatments (Fig. 2b). In all combinations of air temperature and irradiance (CS, CL, WS, WL), R. fruticosus had the significantly lowest biomass per individual plant compared with beech, grown with or without R. fruticosus.

Temperature had no significant effect on the specific leaf area (SLA, m2 g−1) of the B, B + R treatments (Fig. 2c). In the B + R treatment the effect of irradiance on SLA was just above the significance level (P = 0.048). On the other hand, both temperature and irradiance significantly influenced the SLA of R. fruticosus. A significant interaction between these parameters was due to the differential effect of irradiance within each temperature regime; high irrradiance increased SLA of R. fruticosus under low temperature and decreased it under high temperature. Because of this response, SLA of R. fruticosus was significantly lower than that of beech under the CS and WL regimes.

A significant effect of air temperature and irradiance on the root-to-shoot ratios was only found for beech seedlings grown with R. fruticosus (B + R). In the WL growth regime (Fig. 2d) root-to-shoot ratio of beech grown with R. fruticosus was significantly higher compared with beech growing alone. Moreover, in the same growth regime R. fruticosus had the highest root-to-shoot ratio of all plants.

Nitrogen balance

15N uptake  Air temperature had a significant effect on both 15N uptake per plant and specific 15N uptake of beech seedlings grown alone; it resulted in significantly increased values of 15N uptake as temperature increased (Fig. 3a,b). As a result, under the WL regimes beech grown alone had significantly higher uptake rates per plant and per g d. wt compared with seedlings grown in the B + R treatment. R. fruticosus showed significantly higher specific 15N uptake rates than beech of the B and B + R treatments, under all growth conditions (Fig. 3a). However, such an effect was not observed for the plant-based 15N uptake rate of R. fruticosus (Fig. 3b).

Figure 3.

Effects of irradiance, temperature (CS, cold-shade; CL, cold-full light; WS, warm-shade; WL, warm-full light) and competitive interference (B (Fagus sylvatica), beech as control; B + R, beech with Rubus fruticosus; R, R. fruticosus with beech) on specific 15N uptake (a) and 15N uptake per plant (b) of beech seedlings and R. fruticosus plants. Vertical bars indicate ± SE of the mean. Within each treatment, means of the competitive interference treatments are significantly different at a 95% level of significance, when they share no common letter. The effects of temperature (T) and irradiance (L) on each competitive interference treatment (B, B + R, R) were the following: (a) Specific 15N uptake, B: T P = 0.001, L ns, TxL ns; B ± R: T ns, L ns, TxL P = 0.017; R: T ns, L ns, TxL ns (b) 15N uptake per plant, B: T P = 0.004, L ns, TxL ns; B ± R: T ns, L ns, TxL P = 0.021; R: T ns, L P = 0.045, TxL P = 0.043 (ns: P > 0.05).

Total N concentration  The total N concentration on a mass basis in different tissues (stems and leaves) of all interference treatments and growth regimes is presented in Fig. 4(a,b). In stems, no significant effect of either air temperature or irradiance was found. For a given growth regime, R. fruticosus presented significantly higher total N concentration, compared with beech seedlings.

Figure 4.

Effects of irradiance and temperature (CS, cold-shade; CL, cold-full light; WS, warm-shade; WL, warm-full light) and competitive interference (B (Fagus sylvatica), beech as control; B + R, beech with Rubus fruticosus; R, R. fruticosus with beech) on total N concentration of (a) stem, on a mass basis (b) leaves, on a mass basis, and (c) leaves, on an area basis of beech seedlings and R. fruticosus plants. Vertical bars indicate ± SE of the mean. Within each treatment, means of the competitive interference treatments are significantly different at a 95% level of significance, when they share no common letter. The effects of temperature (T) and irradiance (L) on each competitive interference treatment (B, B + R, R) were the following: (a) For stems, B: T ns, L ns, TxL ns; B ± R: T ns, L ns, TxL ns; R: T ns, L ns, TxL ns (b) For leaves on a mass basis, B: T P = 0.001, L P < 0.001, TxL ns; B ± R: T ns, L ns, TxL P = 0.023; R: T ns, L ns, TxL ns (c) For leaves on an area basis, B: T ns, L ns, TxL ns; B ± R: T ns, L P = 0.003, TxL ns; R: T P = 0.01, L P = 0.004, TxL P < 0.001 (ns: P > 0.05).

In leaves of the B-treatment, both air temperature and irradiance had a highly significant effect on the total N concentration: it increased with increasing temperature and decreased with high irradiance. R. fruticosus had the lowest leaf N concentration compared with beech, either with or without R. fruticosus, under the CS and WS regimes.

The foliar N concentration on an area basis is presented in Fig. 4(c). Neither temperature nor irradiance affected the N concentration of the B-treament. In the B + R treatment, increased irradiance resulted in increased N concentration. An interaction was found between the effects of temperature and irradiance on the N concentration of R. fruticosus. The differential effect of irradiance within the temperature regimes was responsible for this; increased irradiance resulted in decreased N concentration under lower temperature and increased N concentration under higher temperature. As a result, the N concentration of R. fruticosus was significantly higher than that of beech at the CS and WL growth regimes, and lower at the CL and WS growth regimes.

Water and carbon balance

δ13C composition Fig. 5 shows the foliar δ13C composition of all interference treatments and growth regimes. For the B treatment, neither temperature nor irradiance had an effect on foliar δ13C. On the contrary, beech growing together with R. fruticosus (B + R) showed 13C-depletion in leaves when temperature increased. However, at high air temperature, the increase of irradiance to the L – light saturating level caused significant 13C-enrichment in the leaves of beech. Comparable light-dependent patterns were also observed for the foliar δ13C signature of R. fruticosus for both temperature regimes: decreased irradiance led to a significant 13C-depletion of leaves. In all growth regimes, apart the WL, leaves of R. fruticosus were significantly 13C-depleted compared with beech seedlings.

Figure 5.

Effects of irradiance and temperature (CS, cold-shade; CL, cold-full light; WS, warm-shade; WL, warm-full light) and competitive interference (B (Fagus sylvatica), beech as control; B + R, beech with Rubus fruticosus; R, R. fruticosus with beech) on the foliar δ13C signature of beech seedlings and R. fruticosus plants. Vertical bars indicate ± SE of the mean. Within each treatment, means of the competitive interference treatments are significantly different at a 95% level of significance, when they share no common letter. The effects of temperature (T) and irradiance (L) on each competitive interference treatment (B, B + R, R) were the following: B: T ns, L ns, TxL ns; B ± R: T P < 0.001, L P = 0.007, TxL P = 0.024; R: T ns, L P = 0.07, TxL P = 0.036 (ns: P > 0.05).

Leaf transpiration and CO2-assimilation rates

Fig. 6 shows the assimilation and transpiration rates in all interference treatments and growth regimes. Assimilation rates of beech seedlings of both interference treatments tended to increase as air temperature and irradiance increased (Fig. 6a). However, the only significant effect was that of air temperature for the B + R treatment. In the WS and WL growth regimes R. fruticosus had significantly lower CO2 assimilation rates, compared with beech, grown with or without R. fruticosus.

Figure 6.

Effects of irradiance and temperature (CS, cold-shade; CL, cold-full light; WS, warm-shade; WL, warm-full light) and competitive interference (B (Fagus sylvatica), beech as control; B + R, beech with Rubus fruticosus; R, R. fruticosus with beech) on assimilation rate (a) and transpiration rate (b) of beech seedlings and R. fruticosus plants, measured shortly before the end of the experiment (18–19 April 2001). Vertical bars indicate ± SE of the mean. Within each treatment, means of the competitive interference treatments are significantly different at a 95% level of significance, when they share no common letter. The effects of temperature (T) and irradiance (L) on each competitive interference treatment (B, B + R, R) were the following: (a) Assimilation rate, B: T ns, L ns, TxL ns; B ± R: T P = 0.001, L ns, TxL ns; R: T ns, L ns, TxL ns (b) Transpiration rate, B: T P < 0.001, L ns, TxL ns; B ± R: T P < 0.001, L ns, TxL ns; R: T P = 0.001, L ns, TxL ns (ns: P > 0.05).

Transpiration rates increased significantly with increasing air temperature for all interference treatments, while irradiance had no significant effect. In most of the cases no significant differences were found between the interference treatments.

Discussion

Previous studies have shown that both the water and the nitrogen balance of young beech seedlings were seriously impaired by competitive interference induced by the early successional, invasive R. fruticosus under water shortage (Fotelli et al., 2001; Fotelli et al., 2002b). However, information regarding the additional effects of air temperature and irradiance on this interference is missing. The influence of these abiotic factors is of particular importance since air temperature is a main component of the climate change scenarios (e.g. IPCC, 2001), and irradiance is greatly altered after opening the forest canopy for promoting the natural regeneration in beech forests (Dertz, 1996; Tarp et al., 2000). Hence, we assessed the effects of varying irradiance and temperature on the nitrogen, water and carbon balance of beech seedlings, under the influence of competitive interference by R. fruticosus, at induced moderate drought.

In general, the presence of R. fruticosus as a strong fast-growing competitor induced no particular patterns of competitive interference on beech at lower air temperature, either together with shade (CS) or with full light (CL). This view is supported by the absence of any significant differences between the control (B) and the applied interference (B + R) for most of the parameters tested (Figs 2–6).

A higher air temperature resulted in higher transpiration rates in beech, compared with the lower-air temperature regimes (Fig. 6), in both B and B + R treatments. Similarly, increased rates of gas exchange due to warming are reported by other studies (Loik et al., 2000; Starr et al., 2000). However, this response of gas exchange to warming did not affect the foliar δ13C of the plants within our study, as also found by Llorens et al. (2003). Contrary to what would have been expected, air temperature had no significant effect on assimilation rates (Fig. 6). This could not be due to a potential plant N-deficiency, given that total leaf N contents of beech (c. 1.8–3.0%) and blackberry (c. 1.6–2.8%) in this study were not lower than the ones found under natural field conditions.

Irradiance is generally assumed to affect gas exchange patterns of beech seedlings (Welander & Ottosson, 1998; Aranda et al., 2004) by increasing CO2 fixation and transpiration with increased irradiance. However, no such effect was found in our study (Fig. 6). Probably, the modulated soil water availability the plants were subjected to (Fig. 1b) leveled the effect of irradiance on gas exchange. This might also be the case for the lack of temperature effect on assimilation rates.

Effects on the competitive interference between R. fruticosus and beech seedlings were observed within the ‘high temperature’ regime, but varied between the two irradiance-regimes (shade vs full light). Under elevated temperature and irradiance R. fruticosus increased its root-to-shoot ratio (Fig. 2d; WL) and decreased its SLA (Fig. 2c; WL), in order to increase its root biomass, and thus water and nutrient absorbing area and to limit water losses from the leaves. The latter one apparently failed, since no transpiration reduction was found for R. fruticosus under the WL regime (Fig. 6b). However, an ability of R. fruticosus to increase its root biomass under unfavorable conditions, such as limited water availability, has been recorded before (Fotelli et al., 2001). Given that water availability was controlled and relatively restricted in all growth regimes, it is indicated that R. fruticosus occupies the strategy of increasing its root biomass under elevated temperature and irradiance, conditions that are adverse for beech (Tognetti et al., 1998; Fotelli et al., 2002a; Fotelli et al., 2003). Under these apparently unfavorable conditions (WL), the foliar biomass increment of beech grown with R. fruticosus was substantially decreased, compared with the CS, CL, WS growth regimes (Fig. 2a). The same was the case for the 15N uptake rates of beech grown with R. fruticosus (Fig. 3a,b) as previously observed by Fotelli et al. (2002b).

On the contrary, under the WL conditions, the foliar N concentration on an area basis was substantially higher in R. fruticosus than in beech (Fig. 4c). Moreover, the specific 15N uptake rate of R. fruticosus was always higher than that of beech (Fig. 3a), in accordance to the findings of Fotelli et al. (2002b). The generally lower 15N uptake rate per plant of R. fruticosus was due to its lower biomass, compared with beech (Fig. 2b). Still, in the WL regime, even though the biomass of beech grown with R. fruticosus (B + R) was c. 4-fold higher than of R. fruticosus (Fig. 2b), its 15N uptake per plant was lower (Fig. 3b). The fact that the initial biomass of the beech seedlings used for this experiment was higher than in other studies (Fotelli et al., 2001; Fotelli et al., 2002b) probably made the negative effect of R. fruticosus on the N status of beech under elevated temperature and irradiance less intensive. Perhaps, the adverse effects of R. fruticosus would have been more pronounced if the unfavorable growth conditions had been applied for a longer period of time. Nevertheless, the findings of Valladares et al. (2002) support the conclusion that under elevated irradiance beech seedlings have lower plasticity, and thus competitiveness against more light-demanding species. In agreement with our observations Coll et al. (2004) also showed that beech seedlings growing together with different grass species are unable to compete effectively for soil N under conditions of high irradiance.

Under elevated temperature, reduced irradiance seems to improve the competitiveness of beech relative to R. fruticosus. Then, the biomass and root-to-shoot ratio of R. fruticosus were the lowest compared with all other growth regimes (Fig. 2b,d-WS). Although specific 15N uptake of R. fruticosus was the highest under the WS regime (Fig. 3a), this could not compensate for its reduced root biomass, resulting in low N uptake per plant, compared with beech (Fig. 3b). Moreover, under these conditions, the significantly lower foliar total N concentration of R. fruticosus, compared with beech (Fig. 5b,c), led to significantly lower assimilation rates (Fig. 6a). Under elevated temperature and shade, the photosynthetic activity of R. fruticosus was rather limited, as indicated by reduced assimilation rates (Fig. 6a) and particularly negative foliar δ13C values (Fig. 6). R. fruticosus probably responded by substantially increasing N allocation to above-ground compartments (Fig. 4a), a major N-storage tissue (Millard & Proe, 1992; Millard & Proe, 1993), in order to enhance its photosynthetic activity. However, this response was, as expected, inadequate for improving the physiological status of R. fruticosus under the WS conditions, since it is an early successional, light-demanding species, contrary to the shade-tolerant beech (e.g. Valladares et al., 2002; Aranda et al., 2004).

In conclusion, competitive interference between beech seedlings and R. fruticosus was enhanced under elevated air temperature, as indicated by changes in their physiological responses. The shade-tolerant beech seedlings appeared to be more competitive when elevated air temperature was accompanied by reduced irradiance. Then, the N status of R. fruticosus was impaired, while its photosynthetic activity and growth were particularly decreased. On the contrary, when both air temperature and irradiance were high, R. fruticosus appeared to be more efficient in resources acquisition, while the pedospheric N acquisition and the N status of beech seedlings were impaired.

The present findings strengthen the hypothesis of Fotelli et al. (2004) derived from field studies that thinning, which induces higher radiation input, adversely affects the physiological performance of the drought-sensitive beech at elevated air temperatures and reduced soil water availability. In addition, we showed here that not only the direct effects of changing abiotic factors – for example water availability (Fotelli et al., 2001; Fotelli et al., 2002b), irradiance and temperature (present study) but also changes in competitive patterns between beech and drought resistant, light demanding species may be responsible for the effects observed in the field (Fotelli et al., 2004). Longer period-experiments would contribute greatly to the simulation of processes taking place during the entire growing season. Still, the present results underlines the need to adapt silvicultural techniques as thinning and weed control to the expected climate changes.

Acknowledgements

M.N. Fotelli thanks the LGF (Landesgraduiertenförderung) of Baden-Württemberg for the financial support during this study. This work was funded by the DFG (Deutsche Forschungsgemeinschaft) under contract numbers SFB 433 and Re515/13–1.

Ancillary