1. The aim of the study was to investigate effects of enhanced UV-B radiation on the balance between biomass production and decay in an ombrotrophic bog which is dominated by one species of Sphagnum (S. fuscum). This paper concerns production.
2. Enhanced UV-B radiation (simulating 15% ozone depletion under clear sky conditions) was applied by means of fluorescent tubes during two growing seasons.
3. In S. fuscum, shoot density, mass relations and length increment over time were measured and productivity was estimated. Pigment concentration, rates of dark respiration and maximum net photosynthesis were recorded.
4.Sphagnum fuscum productivity was not changed by enhanced UV-B radiation while properties determining production were highly influenced although in opposite directions.
5. Height increment was decreased by 20% in the first growing season and by 31% in the second growing season under enhanced UV-B radiation. After two growing seasons spatial shoot density was decreased by 8% by enhanced UV-B radiation. The shoots became stunted as capitulum dry mass and stem dry mass per unit length were increased by 21 and 17%, respectively, under enhanced UV-B radiation.
6. Dark respiration was significantly decreased by 31% after growth under enhanced UV-B radiation.
7. The UV-B induced change in shoot biometry together with the reduced spatial shoot density involve potential long-term effects on peat structure with possible feedback on productivity, decomposition and the strength of the system as a carbon sink.
The depletion of the stratospheric ozone layer and its concomitant enhancement of biologically harmful ultraviolet-B radiation (UV-B, 280–320 nm) reaching the surface of the Earth (Stolarski et al. 1992) has been of increasing political and scientific concern over the last two decades. So far, effects of UV-B radiation on plants have mainly been studied on single plant species with focus on economically important crops. Most investigations have been carried out under artificial light and growth conditions which may overestimate effects by UV-B radiation (for review Caldwell & Flint 1994). Results from such studies are often difficult to apply to ecologically relevant conditions. The few UV-B experiments carried out on an ecosystem level have shown UV-B sensitivity of ecologically relevant variables, such as growth in dwarf shrubs (Johanson et al. 1995), bryophytes (Gehrke et al. 1996; Gehrke, in press) and grasses (Tosserams, Pais de Sà & Rozema 1996), and on the chemical quality of leaf litter and thereby on decomposition of organic matter (Gehrke et al. 1995; Rozema et al. 1997). However, UV-B studies which consider both the productivity and decay from an ecosystem point of view and estimate potential effects on the biogeochemical cycle are missing even though the issue has already been raised (SCOPE 1993). This is the first article in a series which intends to fill that gap.
Ecosystems at high latitudes are supposed to be particularily sensitive to an increase in UV-B radiation because they are adapted to a naturally low UV-B regime (Caldwell, Robberecht & Billings 1980). High latitude ecosystems of particular importance even at a global scale are peatlands. Boreal and subarctic regions contain the largest areas of peatland worldwide (346 Mha, Gorham 1991), storing carbon as organic matter (455 Mt as estimated by Gorham 1991) which amounts to approximately one-third of the total world pool of soil carbon. There is twice as much carbon stored in peatlands globally as is fixed by all terrestrial vegetation within 1 year (Clymo & Hayward 1982). Thus peatlands have been strong carbon sinks but they have the potential to turn into sources in the future because of changing environmental factors. This would have effects on a larger biological and geographical scale and potentially feeds back to the global climate.
The strength of peatlands as a carbon sink is determined by the balance between peat production and peat decay. Peat is mainly produced by bryophytes of the genus Sphagnum. Although vascular plants may add considerable amounts of carbon as roots and their exudates, they do not directly contribute to the long-term accumulation of carbon (Wallén 1992). The bog mosses Sphagnum act as the ecological engineers creating the conditions that lead to peat formation and providing the structure which allows the assemblage to function as a peatland system (van Bremen 1995). Biomass production in Sphagnum is determined by height increment, biomass per unit shoot length and spatial shoot density. The first two variables were shown to be UV-B sensitive in vascular plants (e.g. Fox & Caldwell 1978; Barnes et al. 1993; Tosserams & Rozema 1995) and in the bryophytes Hylocomium splendens and Polytrichum commune (Gehrke et al. 1996; Gehrke, in press). Decay is mainly determined by the microbial assemblage, the substrate quality, the depth of the aerated zone (acrotelm) and the residence time for the peat litter in the acrotelm. The structure and function of microbial communities as well as the substrate quality and subsequent decay are UV-B-sensitive variables (Gehrke et al. 1995; Rozema et al. 1997).
The aim of the present study was to estimate effects of enhanced UV-B radiation on the balance between production and decay in a subarctic bog dominated by one hummock forming Sphagnum species (Sphagnum fuscum). Vascular plants are of low abundance. This paper mainly addresses effects of enhanced UV-B radiation on variables determining productivity of S. fuscum. Additionally, in a mechanistic approach, effects on physiological variables were studied in order to find out where possible changes in growth may originate. A second paper on UV-B effects on peat quality and decay is in progress.
Materials and methods
SITE, CLIMATE AND SPECIES
The experiment was carried out in a subarctic ombrotrophic peat bog close to the Abisko Scientific Research Station (68·35 ° N, 18·82 ° E, 360 m a.s.l.) in northern Swedish Lapland during 1994 and 1995. The mean annual air temperature at Abisko is – 0·8 °C with July being the warmest summer month (11 °C) as calculated from data for 1961–1990. Of the rather low mean annual precipitation (304 mm, mean for 1961–1990) approximately one-third occurs during the summer months June to August.
The study was performed within a 40 m × 20 m area with continuous cover of S. fuscum, sparsely intervowen with two liverworts (Mylia anomala and Calypogeia sphagnicola). The depth of the acrotelm is c. 40 cm. The studied bog consists of S. fuscum hummocks with sparsely scattered Rubus chamaemorus, Drosera rotundifolia, Andromeda polifolia, Empetrum hermaphroditum, Vaccinium microcarpum and Salix spp.
Ten plots, each 20 cm × 20 cm, were randomly chosen. A square frame (18 cm × 18 cm × 25 cm high), giving an effective, i.e. evenly irradiated, area of 10 cm × 10 cm within which biological measurements were made, was erected above each of the 10 plots. Each frame carried four UV-B tubes (Philips TL/4 W) arranged in a square (Gehrke et al. 1996). Five plots were exposed to enhanced UV-B radiation by adding UV-B from the tubes, simulating 15% ozone depletion under clear sky conditions (treatment). The ecologically irrelevant short-wave UV radiation (UV-C, < 280 nm) emitted by the tubes was excluded by cellulose diacetate filters wrapped around each tube. The five plots serving as controls received solar ambient UV-B only. Here wavelengths shorter than 318 nm (virtually all UV-B) emitted by the tubes were cut off by Mylar film. The tubes were preburned for 100 h until they had reached a stable output. Filters were presolarized and exchanged after 30 irradiation hours in the field, resulting in stable transmission properties during use.
UV-B emitted by the tubes was measured with a spectroradiometer (Optronics 742, Orlando, FL, USA). To derive the biologically effective UV-B radiation (UV-BBE) a modification of Caldwell’s generalized plant action spectrum (Caldwell 1971) as parametrized by Thimijan, Carns & Campbell (1978) and normalized at 300 nm was used. The extra UV-B dose required to simulate 15% ozone depletion at Abisko was calculated with a computer model (Björn & Murphy 1985; Björn & Teramura 1993) for clear sky conditions, aerosol level zero and 50% air humidity. The maximum midsummer values calculated by the model (without ozone depletion 4·6 kJ m–2 h–1 UV-BBE; under 15% ozone depletion 5·8 kJ m–2 h–1 UV-BBE) were confirmed in situ by measuring the solar ambient UV-B and the UV-B emitted from the tubes on clear days.
Taking cloudiness into account the extra UV-B given simulated an ozone depletion slightly higher than 15% (e.g. 19% in 1994 according to Björn & Holmgren 1996). Radiation was applied between May and September. The daily irradiation was centred around solar noon and controlled by timers. All four tubes in a frame were switched on simultaneously. Every second week, the daily irradiation duration was adjusted to follow the seasonal change in natural UV-B radiation.
At the start of the experiment (June 1994) samples of S. fuscum were taken 2 cm outside each plot, assuming them to be representative of the adjacent plot. They were analysed for shoot mass relations in order to determine the initial variation among plots assigned to controls and treatment. At the end of the experiment (August 1995) samples were taken only from unshaded spots within a plot in order to exclude shading by vascular plants that influences any UV-B responses in S. fuscum.
Data on vascular plants were sampled in order to describe the studied bog. At the end of the experiment the percentage cover of vascular plants was measured with an area meter after tracing onto transparent overhead film. Above-ground biomass was clipped at the top of the canopy, oven dried (24 h, 80 °C) and weighed.
Shoot density of S. fuscum was measured as number of capitula per unit area. Each plot was divided into 25 subsquares (2 cm × 2 cm) by a grid. The capitula were counted in the same 10 randomly chosen subsquares in each plot both at the start and at the end of the experiment.
Length increment was recorded with external reference marks inserted in unshaded spots, i.e. at a distance from vascular plants. The wires (as described by Clymo 1970) consisted of cranked stainless steel wires, 10 per plot. Measurements (to the nearest 0·5 mm) were made every 4 weeks during the first growing season and every 2 weeks during the second growing season. During the winter season the wires were left untouched in the carpets. At the start of irradiation in the second season, a new zero-point for length increment was established.
Mass relations in shoots were measured in 10 samples per plot comprising three pooled S. fuscum shoots each. The top 0·5 cm (capitulum) was separated from the subsequent stem section. For samples taken before the experiment a defined stem length of 1 cm was cut underneath the capitulum. For samples taken at the end of the experiment the length of the cut stem was given by the mean length increment of the respective plot as measured with the cranked-wire method. Dry mass was determined after freeze drying. All masses were calculated as means of the three pooled shoots per sample.
Shoot biomass increase ΔB was calculated as
were Bstem is the biomass per unit length below the capitulum and ΔL is the mean increment in length in each plot. A correction for a possible change in capitulum biomass (ΔBcap) was applied:
were Bcap 94 is the mean capitulum biomass in each plot as measured before the start of the experiment (1994) and Bcap 95 is the mean capitulum biomass in each plot at the end of the second growing season, i.e. as harvested in 1995. The ΔBcap was then added to eqn 1:
Sphagnum production (g dm–2) during the experimental period, 2 years, was calculated as the mean shoot biomass per plot as derived from eqn 3 multiplied by the mean spatial capitulum density of the respective plot.
Gas exchange was measured in vitro using an Infrared Gas Analyser, IRGA (ADC Ltd, Hoddesdon, UK). The open gas-exchange system, enabling differential readings of the CO2 concentration between air going into and coming out of temperature-controlled cuvettes (flow rate 12 litres h–1), is described by Sonesson (1989). The light source was a metal halogen lamp (Osram HQI, TS 400 W). Five samples were taken per plot, comprising 20 pooled capitula each, and put in nylon trays (5 cm in diameter) in upright position with capitula all level. Each basket was inserted in a Plexiglas cuvette (300 cm3). After 4 min maximum net photosynthesis was measured at ambient CO2, 15 ± 0·05 °C, saturating PAR (900 μmol m–2 s–1, H. Rydin, unpublished data). Immediately afterwards the cuvettes were covered with aluminium foil and after a few minutes dark respiration was measured. After freeze drying, dry mass and pigment concentration were determined. Gas-exchange rates were calculated per unit dry mass and photosynthetic rates were additionally expressed per unit chlorophyll mass. In order to control the water content the capitula were irrigated with rain-water prior to measurements and relative water content was determined by weighing both before and after gas-exchange measurements. The range of optimal water content (i.e. water content enabling maximum net photosynthetic rates) was determined using S. fuscum capitula from outside the experimental area. It was found to range between 380 and 620% of dry mass (calculated as the difference between fresh mass and dry mass divided by the latter). This is low compared to other studies (Rydin & McDonald 1985; Silvola 1990), possibly owing to site-specific or method-specific differences.
Pigment concentration in S. fuscum was determined after gas-exchange measurements. The samples were freeze dried (Leybold Heraeus, type GT2) and homogenized using mortar and pestle. Chlorophylls and carotenoids were extracted in 80% acetone. UV-B screening pigments were extracted in acidified methanol (methanol: H2O:HCl; volume 79:20:1). The extract was centrifuged twice at 1600 g for 10 min (Hettich Universal, type 1200/D) and analysed spectrophotometrically (Shimadzu UV-160 A) at 470·0, 646·8 and 663·2 nm for chlorophyll and carotenoids, and between 280 and 320 nm for UV-B screening pigments. Chlorophylls and carotenoids were calculated on dry-mass basis according to Lichtenthaler (1987). The amount of UV-B screening pigments was expressed in arbitrary units as area under the absorption curve between 280 and 320 nm (AUC280–320) calculated per unit dry mass.
Effects of enhanced UV-B radiation on productivity were tested by one-way ANOVA using each plot as a replicate. Length increment data were treated with a repeated measure ANOVA. All other data on S. fuscum were analysed for UV-B effects with a hierarchic ANOVA, nesting the 10 (mass relations, shoot density) or five (pigmentation and gas exchange) measurements in each of 10 plots within the two treatment levels (ambient and enhanced UV-B). For all variables the results of the statistical analysis are only reported for the treatment. Statistics were performed with the program SYSTAT (version 5·2 for Macintosh, SYSTAT Inc., Evanston, IL, USA). Where necessary data were transformed before analysis but all data are presented without transformation.
The cover of vascular plants was hardly 10% (Table 1). Their above-ground biomass was less than 20 g m–2 (Table 1) which is negligible compared to 170 g m–2 that was reported by Wallén (1986, 1992) for hummocks dominated by S. fuscum on the Stordalen mire 20 km east of Abisko.
Table 1. . Above-ground biomass (clipped at the top of the Sphagnum layer) and cover of vascular plants after exposure of the hummock to ambient and enhanced UV-B radiation during two growing seasons. Mean ± SE, n = 5
The spatial density of S. fuscum capitula (c. 8 cm–2) was not different among plots assigned to control and treatment at the start of the experiment (F1,90 = 0·86, P = 0·36, nested ANOVA). At the end of the experiment the density was significantly decreased by 8% by enhanced UV-B in comparison to ambient UV-B (Fig. 1).
As the season progressed, length of S. fuscum increased approximately asymptotically towards 4·9 ± 0·3 mm and 6·1 ± 0·4 mm under ambient UV-B during 1994 and 1995, respectively (Fig. 2). Under enhanced UV-B radiation length showed a constantly slower increment rate. Total length increment under enhanced UV-B radiation was 3·9 ± 0·2 mm in 1994 and 4·3 ± 0·4 mm in 1995. That corresponds to a significant decrease in annual length increment by 20 and 31% in 1994 and 1995, respectively, compared to ambient UV-B radiation (Fig. 2).
At the start, mass relations within S. fuscum shoots were not different between plots assigned to treatments either in capitulum (F1,90 = 1·0, P = 0·32, nested ANOVA) or stem sections (F1,90 = 1·55, P = 0·22, nested ANOVA). At the end of the experiment capitulum dry mass and stem dry mass per unit length were significantly increased by 21 and 17%, respectively, owing to enhanced UV-B radiation (Fig. 3).
The production of biomass in S. fuscum from the start of the experiment to the end, two seasons later, was c. 1 g dm–2, and it was not different between treatments (Fig. 4).
Maximum in vitro net photosynthesis (NPmax) in S. fuscum grown under enhanced UV-B radiation was significantly increased by 15% when calculated per unit chlorophyll. However, no difference existed when NPmax was calculated on a dry-mass basis (Fig. 5). Dark respiration on a dry-mass basis was significantly decreased by 31% after growth under enhanced UV-B radiation (Fig. 5). During gas-exchange measurements the mean water content was 420 ± 130% (range 210–820%) for samples grown under ambient and 450 ± 130% for samples grown under enhanced UV-B radiation (range 210–760%). Water contents which fell within the optimal range of 380–620% dry mass were observed in 80% of the samples.
Chlorophyll a was significantly decreased by 12% (Table 2) under enhanced UV-B radiation. The strong decrease in the carotenoid concentration (by 25% compared to ambient UV-B) led to a significantly higher chlorophyll carotenoid quotient under enhanced than under ambient UV-B radiation (Table 2). The concentration of UV-B screening pigments was unchanged (Table 2).
Table 2. . Pigmentation (chlorophyll, carotenoids, UV screening pigments) in capitula of Sphagnum fuscum after exposure to ambient and enhanced UV-B radiation during two growing seasons. Mean ± SE, five measurements in each of five plots per treatment, df = 1, 40. F and P values according to nested ANOVA. AUC280–320, area under absorption curve between 280 and 320 nm; chl, chlorophyll
From this study it cannot be proposed that productivity and thus carbon capture from the atmosphere do change directly owing to enhanced UV-B radiation. Variables determining productivity in S. fuscum were affected by enhanced UV-B radiation in opposite directions and may have compensated each other. However, the studied system is dominated by the hummock forming species S. fuscum while bogs often contain a mixture of Sphagnum species forming a mosaic of hummocks and hollows. Responses to UV-B radiation can highly differ in both magnitude and direction between species as shown for vascular plants (Caldwell et al. 1995). Thus extrapolations to more complex peatlands should be made with care.
Peatlands have an immense innate variability in productivity among microsites (Malmer & Wallén 1993) which was reflected in the present study by high standard errors in productivity, no matter of treatment. This is likely to have overruled any effect of enhanced UV-B radiation applied during only two growing seasons. However, the variables determining productivity (spatial shoot density, shoot mass relations and length increment) and indirectly regulating the structure of the deposited peat and thus the carbon balance of the system were considerably affected by enhanced UV-B radiation.
Incoming sunlight is absorbed by the upper 1–2 cm of the peat layer (Clymo & Hayward 1982). Hence, the part of the shoot mostly exposed to solar radiation and thus UV-B radiation is the capitulum where the main initiation of new biomass takes place. The production involves both an apical addition of biomass and a proximal growth in stem length. Length increment of the stem is the process which distributes the biomass along the stem. The decrease in stem length increment owing to enhanced UV-B radiation caused a stunting of shoots and must be the main reason for the increased accumulation of biomass in the capitulum section and increased stem dry mass per unit length in S. fuscum. From a mechanistic point of view the decrease in length increment, which has also been found in many vascular plant species (Barnes, Flint & Caldwell 1990; Johanson et al. 1995) and in the bryophytes H. splendens and P. commune (Gehrke, in press), may be the result of the absorption of UV-B by the growth hormones and their subsequent photodegradation (Tevini, Mark & Saile-Mark 1991).
The decrease in spatial capitulum density may be mediated by the high innate variation in capitulum size combined with the UV-B induced shoot stunting, especially the increased capitulum dry mass. Size increase in already large capitula may have caused smaller capitula to become buried. The possibility that UV-B radiation suppresses branching, as another possible explanation for the lower capitulum density, is in contrast to the UV-B literature (Barnes, Flint & Caldwell 1990).
Even though the negative effect of enhanced UV-B radiation on annual length increment was considerably higher in 1995 than in 1994 it would be premature to claim that UV-B effects are increasing with time. The irradiation and measuring periods in the 2 years were not exactly matching each other. The steep growth phase of spring was unfortunately left out in 1994 (Fig. 2). Between 5 May and 4 July 1995 a considerable part of the annual increment was accomplished during the early growth period (30% under ambient UV-B and 19% under enhanced UV-B). Thus I have to assume that total length increment in my data set for 1994 is underestimated. In fact, the measured total length increment was 19% (ambient UV-B) and 8% (enhanced UV-B) less in 1994 than in 1995. This in combination with the fact that differences between treatments already show up early in the season (after julian day 170, i.e. 19 June 1995) makes it likely that the missing early months in the data set for 1994 have contributed to the lower response to enhanced UV-B radiation measured in the first year. This may also have influenced the estimation of biomass production.
Decreased rates of dark respiration on a dry mass basis found in the capitula in the present study may originate in effects of UV-B radiation on Sphagnum itself or in inhibition of micro-organisms occurring between branches or leaves in the UV-B exposed capitulum layer. UV-B induced increases rather than decreases of respiration rates have been found in vascular plants (Teramura 1983; Musil & Wand 1993; Naidu et al. 1993) while it was shown that UV-B radiation can indeed affect function (decreased respiration) and structure (frequency of species’ occurrence) of the microbial community involved in decomposition, as shown for deciduous dwarf-shrub litter in a subarctic heath (Gehrke et al. 1995). However, assuming that the measured dark respiration is mainly accounted for by Sphagnum tissue, a decrease in the dark respiration rate owing to an enhancement of UV-B radiation implies an increase in gross photosynthesis as the apparent net photosynthetic rate was unchanged. If this would happen during a substantial time of a day it may lead to an increase in carbon gain which, added up over a growing season, may be another explanation, besides the reduction in length increment, for the increased biomass per unit capitulum and stem length.
A reduction in chlorophyll a and concomitantly in total chlorophyll concentration as shown also by other authors (e.g. Day & Vogelmann 1995) may be caused by inhibition of chlorophyll synthesis or degradation of these pigments or their precursors. The decrease in carotenoid concentration is an unusual finding which may also be explained by photodestruction.
Predictions of implications at an ecosytem level are difficult to make because there are numerous numbers and ways of possible interactions. Peat is the product of climatic conditions since the last ice age. Effects of climate changes including an enhancement of UV-B radiation occurring today will be ‘saved’ owing to low decay rates and transferred into deep peat layers with a considerable time delay. An additional time delay is introduced as ozone-destroying compounds we release today take up to 10 years to reach the stratosphere. Thus it may take decades before the long-term effects of recent human activities show up at an ecosystem level and subsequently at a global scale.
This study shows that enhanced UV-B radiation causes changes in shoot biometry and spatial density in S. fuscum. These are key variables determining the size of the pore space system formed between stems, branches and leaves. The macroscopic structure of the peat influences the capillary raise of water into the photosynthetically active capitula and therefore indirectly productivity (Hayward & Clymo 1982). It also determines the hydraulic resistance to the flow of water through the peat which in turn affects the level of the water table. The position of the water table is an important factor determining emission rates of the two major greenhouse gases methane and carbon dioxide (Svensson & Sundh 1992). If the stunting of shoots and the decrease in shoot density induced by enhanced UV-B radiation would persist over a longer time scale feedback effects on the structure and function of bogs dominated by S. fuscum may be anticipated.
This work was carried out under EC contract No ENV4-CT96–0208. I thank B. Wallén for inspiration and advice. He, L.O. Björn, T.V. Callaghan, H. Rydin and R.S. Clymo made valuable comments on this manuscript. L.O. Björn proposed and T. Murath constructed the irradiation system. K. Anton, K. Kilian, J. Lindeberg and Mattias Ståhl helped during field work. The director M. Sonesson and his staff at the Abisko Scientific Rersearch Station provided logistical support. All are gratefully acknowledged.