•Although the most pronounced effects of stratospheric ozone depletion and climate warming probably will occur in polar regions, arctic lichens have not been much studied in relation to climate change.
•Samples of two arctic cyanolichens of the genus Peltigera, exposed in situ to ambient and enhanced UV-B radiation and ambient and increased temperatures, were collected in 2001, 5 yr after the establishment of the experimental set-up. Thallus dimensions and size, coverage of soralia, nitrogen fixation activity and levels of UV-C-absorbing substances were measured.
•Warming had pronounced positive effects on the tridepsides methyl gyrophorate and gyrophoric acid, and unidentified trace substances. However, the combination of enhanced UV-B and increased temperatures did not lead to higher than control levels. Warming reduced coverage of soralia. There were no significant treatment effects on thallus size, dimensions and nitrogen fixation activity.
•UV-B radiation did not to have any adverse effects. The accumulation of tridepsides with warming may be related to increased activity of pathogenic microorganisms or insect herbivores.
Predictions of future polar climate indicate a rise of summer air temperatures by 2–4°C by the year 2100, partly because of natural climatic fluctuations, and partly because of increased emissions of greenhouse gases (Cattle & Crossley, 1995; Weller, 2000). Thus, assessment of the potential interactive effects of various climatic variables such as UV-B radiation and temperature become increasingly important, and long-term experiments on natural community systems are the best method to obtain reliable results (Gwynn-Jones et al., 1999).
The combined effects of enhanced UV-B radiation and temperature on lichen performance have not received much study. Huiskes et al. (2001) and Lud et al. (2001) did not find any differences in photosystem II efficiency and concentrations of chlorophylls, carotenoids and UV-absorbing compounds in two Antarctic green algal lichens that were exposed to various combinations of UV radiation and temperatures. Sonesson et al. (1995) also failed to detect any interactive effects of UV-B radiation and temperature on photosynthetic efficiency in various lichens, both green algal and cyanobacterial.
The purpose of the current study was to assess the impacts of UV-B radiation and temperature on lichens growing in situ in natural middle Arctic tundra vegetation. Two cyanolichens of the genus Peltigera were chosen, because of their abundance in this habitat, and their general abundance in Arctic–alpine vegetation (Vitikainen, 1994). In particular, we focused on thallus morphology, since visible light and UV radiation may affect thallus size and thickness of lichens (Rikkinen, 1995 and references therein), and the levels of UV-absorbing substances in order to test the hypotheses that these compounds accumulate as a function of increased temperature and enhanced UV-B radiation. The nitrogen fixation potential, which was shown to be severely reduced by UV-B radiation in a Peltigera species and in epiphytic arctic cyanobacteria (Solheim et al., 2002) was also measured.
Materials and Methods
The experimental set-up was established in 1996 on the valley floor at 5 m altitude in Adventdalen, Svalbard (78°17′ N, 16°00′ E), c. 12 km east of Longyearbyen, Norway (Gwynn-Jones et al., 1999; Solheim et al., 2002). The site is situated in one of the warmest areas of Svalbard, classified as middle Arctic tundra zone (Elvebakk, 1997). Mean growing season temperatures (June to August) for the study site measured in 1998 and 1999 were 5.9°C and 4.7°C (Solheim et al., 2002). The vegetation is dominated by Salix polaris Wahlenb., Saxifraga oppositifolia L., Bistorta vivipara (L.) S. F. Gray and several mosses. The lichen cover is relatively sparse. Peltigera didactyla (With.) J. R. Laundon was the only lichen that was observed under all frames. Peltigera rufescens (Weis) Humb. was present under most frames. These two Peltigera species were chosen for this study.
Eight metal frames each holding six UV-B fluorescent tubes were assembled 1.5 m above the ground. Four of the frames served as controls in which all the UV-B radiation from the lamps is excluded by window glass. The UV-B treatment represents a simulation of 15% ozone depletion under clear-sky conditions – see Johanson et al. (1995) for details. The ambient and enhanced biologically effective UV-B radiation for a clear midsummer day is 3.0 kJ m−2 d−1 and 3.8 kJ m−2 d−1, respectively (Gwynn-Jones et al., 1999).
Four treatments were created by placing UV-transparent open top chambers (OTCs) in one half of each of the frames. The OTCs are hexagonal structures, constructed from six panels of transparent polycarbonate, and these lead to elevated temperatures by reducing heat loss and acting as shelters (Marion et al., 1997). In 1997, the OTCs increased mean noon temperature during growing season by 3.0°C, while the relative humidity decreased by 16% relative to ambient (Gwynn-Jones et al., 1999). The four treatments are termed +UVB (enhanced UV-B), +T (increased temperature), +UVB + T (combined treatment) and control (ambient UV-B and temperature).
Sampling took place on 20 July 2001 (in the sixth experimental season). From each of the 16 plots (four for each treatment), six thalli of each of the two Peltigera species were collected. In some plots, less than six thalli were encountered, and P. rufescens was absent in some plots. Because of the very low number of thalli of P. rufescens in the OTCs, we were only able to test for the effects of enhanced UV-B radiation on this species.
Nitrogen fixation measurements
All thalli of P. didactyla were cleaned, air-dried and weighted to the nearest 0.0001 g on a calibrated digital balance. Nitrogen fixation activity was measured on two thalli from each plot using the acetylene reduction assay (Stewart et al., 1967). Only P. didactyla was used. The thalli were acclimatized by regularly spraying with distilled water, and floating them on pure water under actinic light for 12 h. Incubation and measurements of acetylene reduction were performed as described by Zielke et al. (2002) with some minor modifications; samples were placed in 10-ml glass vials and incubated with 10% (v : v) acetylene at 22°C for 80–110 min.
Morphological and physiological measurements
The thalli of P. rufescens were cross-sectioned with sharp scalpels under a stereo microscope. Thin slices were cut off 0.3 mm to 0.5 mm from the lobe apices. Slices were placed on microscope slides, mounted with distilled water, and gently covered with a glass coverslip without using pressure. The slides were investigated under a microscope at ×400 magnification. Dimensions of whole thalli and of each of the layers (i.e. the tomentum of the upper surface, the upper cortex, the cyanobiont layer and the medulla) were measured to the nearest ocular unit (1.5 µm). Three lobes from each thallus were analysed. Thallus dimensions of P. didactyla were not measured, as extraction procedures could affect dimensions.
The thalli of P. didactyla were slightly moistened and placed between paper sheets under gentle pressure for 6 h. When dry, they were photographed with a digital camera at 5 cm distance. Pressure was used to reduce three-dimensionality and to obtain good photographs of the thalli so that sorediate and nonsorediate thallus parts were easily discernible. Soredia are granular vegetative propagules produced in circular clusters on the upper surface. The clusters are termed soralia. A 1 mm2 bar was also photographed together with the thalli. The total area (in pixels and mm2) of the thalli and the sorediate thallus parts were estimated using the lasso tools in Adobe PhotoShop 5.0 (Adobe Systems, San Jose, CA, USA).
Content of UV-C-absorbing compounds
There are two varieties of P. didactyla: one that produces detectable amounts of the two depsides methyl gyrophorate and gyrophoric acid (var. extenuata (Nyl. ex Vainio) Goffinet & Hastings) and one that does not produce depsides (var. didactyla) (Goffinet & Hastings, 1995). No lichen substances have been detected in P. rufescens by thin-layer chromatography (TLC) (Vitikainen, 1994). Preliminary analyses by use of high-performance liquid chromatography (HPLC) confirmed that the depsides in P. didactyla accumulate in the soralia with only faint traces in nonsorediate thallus parts. The HPLC analyses showed that P. rufescens is deficient in lichen substances. Therefore, the content of lichen substances in P. rufescens was not studied further.
The lichen substances in P. didactyla were extracted by immersing the thalli in excess acetone for 24 h, followed by another 24 h in a mixture of acetone and methanol, and finally in pure methanol for another 24 h. Extracts were analysed quantitatively by HPLC according to the method described by Bjerke et al. (2002). A slightly modified gradient was used here, as the proportion of methanol was increased from 30% to 72% over 4 min, and subsequently to 79.6% over 11 min. The other solvent, 1% orthophosphoric acid in ultra-pure water, was decreased accordingly. Pure substance of gyrophoric acid was supplied by colleagues at the University of Valparaíso, Chile, but no pure samples of methyl gyrophorate were available. The pure substance was used to obtain a linear standard curve (r2 = 0.999). Since the UV spectra and the absorption coefficients of gyrophoric acid and methyl gyrophorate are almost identical (Huneck & Yoshimura, 1996), the gyrophoric acid standard curve was also used to estimate absolute concentrations of the other tridepside. It is reasonable to assume that the levels presented are close to ‘true’ levels, and that the deviation from the ‘true’ levels is the same for all samples.
The differences between the four treatments were tested with a fixed one-way or two-way anova design and with Tukey multiple comparison tests using the s-plus 6.1 statistical package for Windows (Insightful Corp., Seattle, WA, USA). Logarithmic or arcsine transformation of data was performed in some cases in order to obtain normality and reduce heterogeneity. Values for P lower than 0.05 were considered significant. The number of replicates were equal to number of frames with lichens per treatment (n = 4 for P. didactyla, n = 3 for P. rufescens).
Thallus dimensions of Peltigera rufescens
The thickness of the four thallus layers of the lichen did not vary much within and between treatments (Fig. 1a). For all four layers the mean thickness was slightly greater in samples exposed to ambient UV-B radiation, but the differences were not significant for any of the cases. The mean thallus thickness was also slightly greater in the ambient UV-B treatment (Fig. 1b), but still not significant (P = 0.22).
Thallus size, development of soralia and nitrogen fixation activity in P. didactyla
The analyses of digital photographs show that the thalli that grew under enhanced UV-B radiation tended to be smaller than those grown under ambient UV-B radiation (Table 1), but the difference is not significant (P = 0.09). The effect of temperature on the coverage of soralia was significant (Table 1). There were no significant effects of UV-B treatment. The acetylene reduction activity varied considerably within treatments, and no temperature or radiation effects were detected (Table 1).
Table 1. Thallus size, thallus area covered by soralia, nitrogen fixation activity and ratio of varieties of Peltigera didactyla
+UVB + T
UVB × T
Values are expressed as a mean ± standard error for the three former variables and as back-transformed mean with lower and upper 95% confidence limits (L1 and L2) for the ratio data (n = 4). anova results are expressed as F-values with P-values in brackets. The treatments are enhanced UV-B (+UVB), increased temperature (+T), combined treatment (+UVB + T) and ambient UV-B and temperature (control).
Thallus size (mm2)
88.7 ± 5.5
76.7 ± 11.1
87.1 ± 11.9
65.2 ± 6.2
Soralia (% thallus size)
20.5 ± 2.9
20.6 ± 2.2
15.3 ± 1.6
14.6 ± 1.7
Ethylene (µmol g−1 h−1) var. didactyla (% of all thalli)
1250 ± 302
852 ± 237
1000 ± 308
1096 ± 308
< 0.01 (0.99)
Chemical analyses of P. didactyla
The HPLC analyses showed that both varieties are present in the plots. The depside-containing variety (var. extenuata) was most common in all treatments, but the ratio varied considerably. More depside-deficient thalli were found under ambient UV-B radiation than under enhanced UV-B radiation, but the mean ratio difference was not significant (P = 0.09, Table 1). The two depsides, methyl gyrophorate and gyrophoric acid, were detected in all specimens of var. extenuata. Their UV spectra are almost identical (Fig. 2a), both having maximum absorption in the UV-C range of the spectrum. Four additional unidentified substances were detected in most specimens. The peaks of these substances were small and sometimes absent, and all these substances absorb mostly in the UV-C range (Fig. 2b). The correlation between the levels of gyrophoric acid and methyl gyrophorate is high (r2 = 0.91), whereas no significant correlation between ethylene production and chemical content was found (r2 = 0.03).
Treatments had identical effect on the concentration of both tridepsides (Fig. 3). Data are presented as dry weight per area of soralia. No significant differences were found between control, +UVB and +UVB + T. However, the levels of the two depsides in thalli grown under increased temperatures were significantly higher than in the three other treatments, and there were significant interactive treatment effects. Mean levels of methyl gyrophorate and gyrophoric acid under increased temperature are 6.7 times and 22.3 times higher than control levels, but as seen from the confidence intervals, the variances within the treatment group were large. anova results for the chemical data expressed as a function of thallus area or weight did not deviate from those presented in Fig. 3. Mean concentrations of methyl gyrophorate range from 0.36% w : w (control) to 1.50% w : w (+T). The same values for gyrophoric acid are 0.02% w : w and 0.15% w : w. The levels of the four unidentified trace substances varied between treatments in a similar manner as the two tridepsides (data not shown).
This study was undertaken to assess separate and combined effects of UV-B radiation and warming on various physiological and chemical performances in arctic lichens. Some studies on the effects of UV-B on lichens have been performed in situ in northern boreal (also called ‘subarctic’) and alpine regions (Sonesson et al., 1995; Bachereau & Asta, 1997; Solheim et al., 2002), but the current study is the first of its kind on arctic lichens. It is also the first long-term study (> 4–5 yr), except for a study on P. aphthosa that had been exposed to enhanced UV-B radiation for 8 yr (Solheim et al., 2002).
Despite growing under different conditions for several years, no effects on thallus layer dimensions of P. rufescens were detected. Peltigera rufescens has a tomentose cortex type. Studies on vascular plants have shown that pubescent leaves are more effective in reflecting longer wavelengths than UV radiation (Holmes & Keiller, 2002). Thus, if pubescence in Peltigera has a similar property, it is not surprising that the thickness of the tomentum was unaffected by enhanced UV-B radiation. Thallus dimensions are probably more affected by water than by radiation, because thin-lobed foliose lichens have less water storage capacity and are therefore more desiccation-sensitive than thick-lobed lichens (Snelgar & Green, 1981; Green & Lange, 1991).
Peltigera didactyla responded to enhanced UV-B radiation by developing slightly smaller (albeit not significantly smaller) thalli than under ambient UV-B radiation. Vascular plants and bryophytes often respond to enhanced UV-B radiation by reduced growth of leaves and stems (Sonesson et al., 1996; Gehrke, 1999; Searles et al., 2001), and a similar response in lichens could therefore be expected. A lower proportion of the upper surface was covered by soralia under increased temperature than under ambient temperature. One might assume that this effect was coupled with the smaller thalli observed under enhanced UV-B radiation, but if so, the interaction should have been significant, which it was not. The initiation of production of soredia in P. didactyla is accompanied by cessation of lobe growth (Stocker-Wörgötter & Türk, 1990). Maximum growth rates of lichens in polar and alpine environments generally occur during the first half of summer (Benedict, 1990; Crittenden, 1998; Hovenden, 2001). A possible explanation is that the increased temperatures in the OTC plots prolonged the period suitable for lobe growth, thus postponing the development of soredia, whereas, under ambient temperatures, the lobe growth rate had already been reduced considerably in the period before harvest.
Our measurements of nitrogen fixation activity failed to detect any significant effects of UV-B radiation and warming. These results contrast with those from the same locality by Solheim et al. (2002) in which the nitrogen fixation activity of cyanobacteria epiphytic on moss leaves was found to be reduced by 50% by enhanced UV-B radiation. Solheim et al. (2002) also found a 50% reduction in the nitrogen fixation activity of P. aphthosa that had been exposed to enhanced UV-B radiation for 8 yr in a northern boreal site. Peltigera aphthosa differs from P. didactyla by having a green algal primary photobiont and cyanobacteria in external cephalodia, by not producing vegetative propagules, and by being richly fertile (Vitikainen, 1994). External cephalodia of Peltigera species often lack the protection served by the upper cortex (Vitikainen, 1994), whereas the cyanobiont layer of P. didactyla has this protection and is therefore exposed to lower levels of UV-B radiation than the cyanobacteria of P. aphthosa in areas where these two lichens grow together. The moss leaves render some protection to the epiphytic cyanobacteria, but this protection is probably not as efficient as the protection provided by the upper cortex of P. didactyla, since moss leaves are more transparent than lichen surfaces. These distinguishing characters may help to explain why the results of the current study and of the study by Solheim et al. (2002) differ.
Goffinet and Hastings (1995) indicated that P. didactyla var. extenuata has a more northern distribution range than var. didactyla, and this was evident from the study site, in particular under enhanced UV-B radiation. The two varieties are considered as genetically different entities (Goffinet & Hastings, 1995). Thus, it is unlikely that a change in climatic conditions initiate the production of tridepsides in var. didactyla. Based on the present data, it is not possible to say whether var. didactyla is less adapted to UV-B radiation than var. extenuata. If the levels of the tridepsides had increased with enhanced UV-B radiation, we could have assumed that some UV protective mechanisms are inherent in var. extenuata, and not in var. didactyla, but as seen in Fig. 3, UV-B radiation did not cause any increase in tridepsides. Furthermore, as seen in Fig. 2, none of the acetone-soluble substances in var. extenuata absorb efficiently in the UV-B range of the spectrum. Instead, warming had a pronounced positive effect on the accumulation of methyl gyrophorate and gyrophoric acid, and also on the levels of the unidentified substances. However, there was no significant increase in tridepsides when lichens were exposed to both enhanced UV-B radiation and warming. Thus, the triggering effects of warming are inhibited by enhanced UV-B.
Depsides have been shown to reduce the activity of microfungi, bacteria and insects, and it has therefore been suggested that the principal role of depsides is to the reduce infections and herbivory (Huneck, 1999; Kumar & Müller, 1999; Müller, 2001). It is possible that the activity of aggressive organisms increased with increasing temperature (Harvell et al., 2002), but that their activity under enhanced UV-B was low either because of avoidance or because of direct damage (Paul, 2000; Björn, 2002). Thus, the increase in methyl gyrophorate and gyrophoric acid could be correlated with increased microbial, microfungal and microfaunal activity. An alternative interpretation is that the tridepsides are storage products resulting from increased photosynthetic activity, which is correlated with increased temperatures. If that is the case, the photosynthetic activity of P. didactyla is significantly reduced by UV-B radiation. However, there are no clear indications of reduced photosynthesis, although not measured directly. For example, nitrogen fixation activity, which often is correlated with photosynthetic activity (Nash & Olafsen, 1995), was not reduced by enhanced UV-B radiation.
In summary, no adverse long-term impacts of enhanced UV-B radiation were detected, whereas increased temperatures directly or indirectly affect the synthesis of lichen depsides and the production of soralia.
The facilities of enhanced UV-B radiation of the vegetation at the site in Adventdalen were started under the European Community Contract EV5V-CT910032, and we are grateful for the support of the scientists involved and being able to continue the UV-B treatments since 1998. We thank our colleagues at the University of Tromsø, Arve Elvebakk, Arild Ernstsen and Espen Hansen, for providing assistance and for sharing their knowledge, opinions and literature. Wanda Quilhot and Cecilia Rubio, University of Valparaíso, kindly provided a pure sample of gyrophoric acid.