Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Importance of spruce swamp‐forests for epiphyte diversity and flora on Picea abies in southern and middle boreal Finland

Importance of spruce swamp‐forests for epiphyte diversity and flora on Picea abies in southern... Kuusinen. M. 1996. Importance of spruce swamp-forests for epiphyte diversity and flora on Piceu ubies in southern and middle boreal Finland. - Ecography 1 9 41-51. The epiphyte (lichens and bryophytes) species richness, diversity and composition on basal trunks of Piceu ubies in spruce swamp-forests were compared to adjacent mesic forests on mineral soil in two southern and two middle boreal sites in old-growth forest patches in Finland. The sampling was carried out along four line transects parallel to swamp-forest margin: I ) in the spruce swampforest. 2) at the swamp forest margin, 3) on mineral soil c. 10-20 m OH the swampforest margin and 4) on mineral soil at least 50 m OHthe swampforest margin. In the two southern boreal sites there was a decreasing trend in the average species number per tree from the trees in swamp-forests (21-25 species) to the trees on mineral soil (17-18 species), whereas in the two more humid and virgin middle boreal sites a similar trend was not detected (25-28 species on all trees). There were no major diHerences in the epiphyte flora between the locations or study sites; the dominant species occurred on nearly all sample trees. The crustose lichens Cliusromum leprostun, Arrliuniu leuropellueu and Lecunuctis ubierinu were the most common species that were mainly confined to the swampforests in the southern boreal sites, C.leproswn also in the middle boreal sites. In addition. several rare species occurred exclusively on the swampforest trees. Lecunucris ubietinu extended significantly higher on the swampforest trees than on the trees on mineral soil in the southern boml sites. The spruce swampforests proved to be one of the most important habitats for maintaining the epiphyte diversity in the boreal forest landscape. hf. Kuusinen. Dept of Ecology und Systemutics. P.O. Box 47. Helsinki. Finland. FIN-OOO14 Uniu. oJ' Fire refugia areas with long continuity of old-growth forest are among the most important habitats in preserving natural species diversity in the boreal forest landscape of Fennoscandia (Esseen et al. 1992, Haila 1994),. Spruce swamp-forests have often been considered as such potential fire refugia (Zackrisson 1977, Zackrisson and Ostlund 1991, Sj6berg and Ericson 1992, Esseen et al. 1992, Angelstam and Rosenberg 1993, Hiirnberg et al. 1999, although some contrasting opinions have recently been presented (Segerstrbm et al. 1994, Oldhammer 1994). Hbrnberg et al. (1995), who studied the long-term continuity of ten old-growth spruce swamp-forest stands in northern Sweden, concluded that swamp-forests were less frequently affected Accepted I2 July 1995 Copyright 0 ECOGRAPHY 1996 ISSN 0906-7590 Printed in Ireland - all rights reserved ECOGRAPHY 1 9 1 (1996) by fire than surrounding forests on mineral soil. Some of their study areas had even been untouched by fire since the accumulation of peat began several thousand yean ago. The human impact was also slight in their study areas. Thus, the spruce swamp-forests may have exceptionally Long stand continuity in the boreal forest landscape and the probability for survival of species requiring long continuity should be higher in swamp forests than in adjacent mineral soil stands. Drainage has already destroyed a major part of Finnish spruce swampforests. Particularly small, thinpeated patches surrounded by mesic forests on mineral soil have been extensively drained. Eurola et a]. (1991) estimated that only c. 25% of the original spruce swamp-forest area (almost I million’ha) was still virgin in southern and central Finland. They considered several types of spruce swamp-forests to have become threatened or rare in Finland. Sjiiberg and Ericson (1992) have emphasized the importance of spruce swampforests and other forested wetlands for various organisms in boreal Fennoscandia. For example, a large number of moisture demanding cryptogams, both lichens and bryophytes, are confined to these habitats. In addition, several threatened species in Finland inhabit spruce swampforests (Rassi and Viisiinen 1987). Although several aspects of the floristic and faunistic composition of swampforests have been studied fairly well in Fennoscandia (Sjdberg and Ericson 1992). the epiphyte communities have received little attention. Only Koskinen (1955) and Somermaa (1972) have presented data on the epiphyte flora of swamp forests in their extensive studies in central Finland and Estonia. The aim of this study was to compare epiphyte species diversity, richness and composition between old Picea d i e s trees 1) in spruce swampforests, 2) at the margin of the swampforest and 3) in adjacent mesic forests on mineral soil. Two particular hypothesis of epiphyte species diversity and richness were tested: 1) whether diversity and richness are higher in the spruce swampforest than in the adjacent mesic forest on mineral soil, or 2) the ecotone between the spruce swamp forest and the mesic forest has the highest diversity and richness due to the presence of species characteristic of both habitats. Epiphyte species mainly confined to spruce swampforests o their margins were particularly r searched for. Study sites were selected in both southern and middle boreal Finland to detect geographical differences in species diversity, richness and composition. This study is part of a project searching for the keystone habitats for epiphyte diversity in the oldgrowth boreal forests of Finland. The spruce swamp-forest stands were small, c. I ha patches, in slight depressions surrounded by forests on mineral soil (Fig. 2). The peat layer of the spruce swampforests was generally thin ( e l m) and true forest species were common in the field layer. The ground layer was mostly covered by Sphagnum spp., especially S girgensohnii. Instead of the term “spruce . swampforest” defined by H6rnberg et al. (1995), the term “spruce mire” has often been used of these habitats in the Finnish literature (Eurola et al. 1984, 1991). The most common spruce swamp-forest types were the thin-peated spruce (KgK) and herb-rich forests (LhK) as well as Vaccinium myrtillus (MK) and Equisetum syluaticwn spruce mire (MkK) with some more herbrich patches (for the Finnish mire type classification and the type symbo1s;see Eurola et al. 1984). The spruce swampforests of the two southern boreal sites Study sites Four study sites were selected in old spruce-dominated forest areas, two (1, 2) in the southern boreal and two (3, 4) in the middle boreal zone (Fig. 1). The southern boreal sites Evo 1 and 2 were located c. 1 km apart from each other in a c. 300 ha old forest patch and the middle boreal sites Kuhmo 3 and 4, c. 4 lun apart from each other in c. 100 ha and c. 80 ha patches of old-growth forest. Managed forests, clear-cuts and fires surrounded the forest patches. In the southern boreal sites some human impact, such as signs of old selective logging, were visible, but the two middle boreal sites were of primeval nature. Each study site included part of a spruce swampforest stand and its margin as well as part of adjacent mesic forest stand on mineral soil. The total area of the sites was 1.5-3 ha. Fig. 1. The location of study sites. The thick lines show the borders between vegetation zones (Ahti et al. 1968): A = hemiboreal. B = southern boreal, C = middle boreal, D = northern boreal. ECOGRAPHY 1 9 1 (19961 Fig. 2. General view of ecotone between spruce swampforest and mesic forest on mineral soil close by the southern boreal site Evo 1. were slightly drained in the late 19th century. The dominant forest site types on the mineral soil were the herb-rich Oxalis-Myrtillus (OMT) and the mesic M yrtillus type (MT) in the southern boreal and the corresponding Geranium-0xalis-Myrtillus (GOMT) and Vuccinium-Myrtillus type (VMT) in the middle boreal sites (for the Finnish forest site type classification, see Kalela 1961). The altitude of the two southern boreal sites was c. 160 m a.s.1. and the middle boreal sites c. 260 m a d . The maximum difference in altitude within each study site was between 5 and I5 m. The mean annual temperature is c. +4"C in the southern boreal sites and c. 1°C in the middle boreal sites (Alalarnmi 1987). The coldest months are January and February with mean temperatures of c. -8°C in the southern and c. - 11°C in the middle boreal sites; the warmest month is July with mean temperatures of c. 17°C and c. 1516°C. respectively. The mean duration of the growing season (number of days with mean temperature exceeding +5"C) is c. 165 d in the southern boreal sites and c. 140 d in the middle boreal sites. The mean annual precipitation is 650-700 rnm in all sites, the wettest months being July and August and the driest ones February and March. The mean maximum snow depth in forests is c. 50 cm in the southern and c. 80 cm in the middle boreal sites. The most significant climatic difference between the southern and middle boreal sites is the higher humidity, measured as the difference between precipitation and evaporation, in the latter sites. This value, counted from the disappearance of snow in spring to the end of July, is c. -60 mm in the southern and c. +40 rnm in the middle boreal sites (Alalammi 1987). Methods Sampling The field work was carried out in the summers of 1993 and 1994. The sampling was performed along four 100-200 m long line transects parallel with the spruce swampforest margin: 1) in the spruce swampforest c. 20 m off the swampforest margin. 2) at the swamp f forest margin, 3) c. 10-20 m o f the swampforest margin on mineral soil and 4) at least 50 rn and up to 200 m o f the swampforest margin on mineral soil. f Eight spruces closest to random points were sampled along each of the four transects. Only straight, living trees with a diameter at breast height (DBH) of at least 18 cm were sampled. Trees with plenty of twigs below two meters height on the trunk were also avoided. The Occurrence of all lichen and bryophyte species on the basal trunk < 2 m in height was recorded on the sample trees. The structure of the epiphyte f o a was lr surveyed more in detail by placing a rectangle made of transparent plastic on each tree above 80 cm in height, on both the south and north sides of the trunk. The 20 x 50 cm rectangle was divided into 100 2 x 5 cm subunits, and a single point was randomly placed in each of these. The number of points hit by each species were counted and these values were used as an estimate of the species' cover percentage on the sample plot. The maximum height of the continuous cover of the CNStose lichen Lecanactis abietina was also measured on each sample tree. Specimens of difficult crustose species were collected for identification with microscope and standard thin layer chromatography (White and James 1985). ECOGRAPHY 1 9 1 (1996) Voucher specimens were deposited at the Botanical Museum of Univ. of Helsinki (H). DBH and height of each sample tree were measured. The canopy cover was estimated in lO%-classes adjacent to each sample tree. The distance from the sample tree to the three closest living trees (DBH > 4 cm) was measured to obtain an estimate of stem density. Age of two or three sample trees on each of the four transects and in each site were measured with an increment borer. Bark samples were collected from the trunks at 0.5-1.5 m above ground outside the sample plots. Three grams of bark was soaked in 37 ml of distilled water for 24 h and the pH value of the solution was measured with a standard pH meter. Data analysis Species diversity at each location was calculated with the Shannon-Wiener diversity index (H’). The cover estimates on the south and north sides of the trunk were combined for each species. The index was calculated using the formula swampforest stands and forest stands on mineral soil were in tree age and stem density (Table I). Age showed a decreasing gradient from the swamp-forest to the t m s on mineral soil in both the southern boreal sites and the middle boreal site 4. The swampforest stands tended to be slightly more open than the forest stands on mineral soil. This was reflected by the lower crown cover and higher distance between t m s in the swamp-forest stands of sites 2, 3 and 4 (Table 1). The sample trees were thinner and shorter in the middle boreal than in the southern boreal sites: the mean DBH was 32 cm and tree height 27 m in the southern boreal sites but only 25 cm and 19 m in the northern boreal sites (Table I). However, the middle boreal trces were generally older than the southern boreal ones: the oldest middle boreal swamp forest tms in the data approached 300 yr while the youngest southern boreal ones were < 100 yr old. The acidity of the bark was slightly >pH 3 on most of the sample trees. The bark pH of the sample trees on site 1 was on an average somewhat lower than on the other sites. H I =i- I pi Inp, Species richness and diversity where pi is the proportion of ith species of total cover. The jackknife procedure proposed by Zahl (1977) and Routledge (1980) was used to estimate the diversity indices and to attach confidence limits to these estimates. Statistical comparisons between the four locations were performed with either the parametric one-way analyses of variance (ANOVA) or the non-parametric Kruskal-Wallis test. Pairwise multiple comparisons between the locations were performed with Tukey’s HSD test using 5% significande level. The significance of the differencesin the occurrence of individual species were tested with the Fisher’s exact test for 2 x 2 tables. Trees on the spruce swampforest and at the swampforest margin were compared to all the trees on mineral soil. The presence and absence of each species formed the other dimension of the table. Data from the two southem boreal and two middle boreal plots were combined in this analyses to accomplish larger sample sizes. All the analyses were carried out with Systat for Windows (Wilkinson et al. 1992). The nomenclature follows Santesson (1993) for lichens and Koponen et al. (1977) for bryophytes. A total of 48 lichen and 12 bryophyte species were found on the 64 southern boreal trees and 60 lichen and 12 bryophyte species on the 64 middle boreal trees (Table 2). In addition, two conspicuous non-lichenized fungi, Arthothelium scandinavicum and Chaenothecopsis viridiulba, were included in the data. In the two southern boreal sites the mean species number per tree was significantly higher on the swampforest trees than on the t m s on mineral soil, whereas a similar trend could not be detected in the middle boreal sites (Fig. 3). The mean species numbers per tree wen generally higher on the middle boreal than on the southern boreal trees: the mean species number was 27 on the middle boreal but only 19 on the southern boreal t m . This difference was largest on the trees on mineral soil. The total number of species also showed a decreasing trend from the swampforests to the forests on mineral soil especially in the southern boreal sites (Table 2). The observed differences were mainly in the lichen flora, whereas the bryophyte species numbers varied less (Fig. 3, Table 2). The jackknife estimates of the Shannon-Wiener diversity index were also somewhat higher on the swampforest trees than on the trees on mineral soil in both southern boreal sites and the ResnltS middle boreal site 3 (Fig. 4). The large variation in the point frequency data, however, complicates the Sample tree and stand characteristics detection of any significant pattern in species diverThe most significant differences between the spruce sity. ECOGRAPHY 19:l (1996) Table I . Characteristics of the sampled Piceu uhies trees in each of the four study sites. Locations of the trees are from the left to right column: 1 ) spruce swamp-forest, 2) swamp-forest margin, 3) mineral soil c. 10-20 m OH the swamp-forest margin and 4) mineral soil at least 50 m OH the swamp-forest margin. Comparisons between locations: I = ANOVA, = Knrskall-Wallis test. The last column shows probabilities for no difference between the locations. N = 8 for each measurement. except for tree age. Swamp EVO I (southern boreal) DBH (cm), mean f S.D. Tree height (m), mean f S.D. Acidity of bark (pH), mean f S.D. Mean distance to 3 closest trees (m). mean f S.D. Canopy cover (%), median Tree age (yr), range (N) EVO 2 (southern boreal) DBH (cm). mean f S.D. Tree height (m). mean f S.D. Acidity of bark (pH), mean f S.D. Mean distance to 3 closest trees (m), mean f S.D. Canopy cover (%), median Tree age (yr). range (N) KUHMO 3 (middle boreal) DBH (cm). mean f S.D. Tree height (m), mean f S.D. Acidity of bark (pH), mean f S.D. Mean distance to 3 closest trees (m). mean f S.D. Canopy cover (YO). median Tree age (yr), range (N) KUHMO 4 (middle boreal) DBH (cm). mean f S.D. Tree height (m). mean f S.D. Acidity of bark (pH), mean f S.D. Medn distance to 3 closest trees (m). mean fS.D. Canopy cover (YO). median Tree age (yr), range (N) 32 f 5 26 f 2 3.1 f0.1 3.0 f0.7 80 145-164 ( 3 ) 35f6 28 f 3 3.3 f0.2 4.2 f0.7 65 145-158 (3) 27 f 5 2Of2 3.4 f0. I 3.3f 1 . 1 60 139-175 (3) 29 f 7 19f2 3.520.1 3.3 f 0.7 50 200-282 (2) Margin 28f6 24 f2 3.1 f 0.2 3.1 f 1.1 75 126-160 (2) 34f7 28 f 2 3.3 f 0 . l 3.9 f 0.9 70 130- 148 (2) 24f4 19f I 3.5f0.1 3.4 f0.8 55 143-172 (3) 26 k 6 19f3 3.3 f 0.2 2.5 f0.6 60 185-200 (2) 10-20 m 50 m off the swampforest margin) are marked with M in the pattern column. The significance of these patterns were tested with the Fisher's exact test for 2 x 2 tables. Significance values: three symbols = p < 0.001. two symbols = p < 0.01, one symbol = p < 0.05; N = 32. Bryophytes are marked with an asterisk. Species Hypogymnia physodes Lepraria spp. Parmeliopsis ambigua Cladonia coniocraea etc. Cladonia cenotea Panneliopsis hyperopia Chaenatheca chrysocephala Pleuro:ium schreberis Pilidiwn pulcherrimwn8 Clahnia digiiaia Plagiotheciwn Iaeium' Platismaiia glauca Lmospora elatina Ochrolechia microsiicioides Bryoria capillarb M.vcobIasr~rsanguinarius Vulpicida pinastri Bryoria fuseucens Dicranwn scopariwn' i+lycoblastus ofinis Dicranwn furccsecns8 Ochrolechia an&ogyna Lecidea nylanderi Micarea prasina Usnea spp. Arthothelium scandinavicum Cladoniafimbriaia Chaenotheca subroscida Lecanuctis abieiina Cliostomum leprosum Brachyiheciwn spp.+ Alectoria sanneniosa Caliciwn uiride Lecidea albofuscescens Ceiraria chlorophylla L0pho:ia longidens' Lecanora hypopta coll. Micarea melaena Caliciwn glaucellwn Arthonia leucopelhea k i a k a Iepranoides Chaenotheca irichialis Pertusaria boreal& Mycoblarm alpinus Arthonia incanrota Biaiom efirescens Japewia subaurvera Lecanora circumborealis Biaiora gyrophorica Cypheliwn inquinans Orihodicramun monimum* Dimerella pineii Hylocomium splendens* Ochrolechia albcflawscem Pohlia nutans' Chaenotheca sternonea Hypocenomyce scalaris Hypogyrnnia tubulosa Lophocolea heierophylla' Pertusaria pupillaris Bryoria furcellata Chaenothecaferruginea Lepidoria reptans' Micarea denigraia Swamp Margin Evo 1-2 (southern boreal) Kuhmo 3-4 (middle boreal) 10-20 m >50 m Pattern Swamp Margin 10-20 m >M m Pattern 1 6 16 1 6 1 6 16 I5 16 13 13 16 1 6 12 12 IS 16 I5 II 1 4 1 6 16 16 16 14 16 13 16 14 I2 9 16 16 16 1 6 1 6 16 I5 I5 14 IS 1 6 I5 1 6 IS 16 sss MM I5 6 I5 II 7 3 sss ss 16 I5 13 1 0 14 14 7 13 8 1 6 I2 14 7 6 9 14 9 MM MM MM MM a a sss a a 6 II 5 7 MMM ss 2 2 I 3 Table 2. (cont.) Species Swamp Margin 10-20 m Evo 1-2 (southem boreal) >50 m Pattern Swamp Margin Kuhmo 3-4 (middle boreal) 10-20 m >50 m Pattern I Varicellaria rhodocarpa Bryoria jremontii Chaenotheca furfuracea Imshaugia aleurites Parmelia sulcata Pertusaria amara Pertusaria coccodes Pseudevernia furfuracea Ramalina thrausta Tetraphu pellucida' Arthonia didyma Barbilophoria attenuata' Caliciwn pannun Chaenotheca Iaevigata Charnothecopsis viridialba Cladonia cornuta Cladonia pleurota Lecanora expallens Lecidea erythrophaea Lopadiwn disciforme Sanionia uncinata' Number of lichens Number of bryophytes Total number of species I I cantly higher on the swampforest trees than on the trees on mineral soil in the two southern boreal sites (Fig. 5). In swamp-forests this species may completely veil the basal trunks of spruces up to 3 m height while it is often absent on spruces on mineral soil. This southern boreal species did not occur in the middle boreal sites. Species richness and diversity The numbers of lichen species observed in this study were relatively high for Picea d i e s . especially in the middle boreal sites (60species on 64 trees). For example, HyvHrinen et al. (1992) found only 38 lichen species on over 100 spruces and Sbmermaa (1972) reported 74 lichen species on spruce in her extensive data for hemiboreal Estonian forests. Koskinen (1955) reported a total of 154 lichen species on Picea d i e s in his study area in central Finland. This very high species number is, however, not quite comparable with my data. His data had been collected from a very large number of trees (though the figures are not given) in various habitats and includes species on both the whole trunk and branches. Furthermore, several of his species are of doubtful taxonomic status, such as many Biatora spp. and Umea spp. Kuusinen (1994a, b) studied the lichen species numbers per tree on Salix caprea and Populus tremula using the same methodology and partly the same areas as in this study. The lichen species numbers per tree were generally higher on Salk caprea (mean 22) and lower on Populus tremula (mean 13) than on Picea abies. ECOGRAPHY 1 9 1 (1996) There are only few estimates of the epiphyte species richness in wet boreal forests. According to Somermaa (1972) alder swamps had the highest epiphytic lichen richness of all Estonian forest types. Koskinen (1955) did not separate thin-peated spruce swampforests of mesic forests in his data, but wet spruce swampforests and pine bogs werc the most epiphyte species poor habitats in his data. He concluded that the low species numbers were mainly due to the scarcity of trees, high light intensity and drought. These habitats are thus quite different from my study sites which had a dense tree stand and closed canopy. Two prominent factors behind the observed high epiphyte species richness and diversity in the spruce swampforests could be: 1) the long continuity and 2) the humid microclimate of the swampforest stands. Forestry and fires, probably also storms, may have less frequently and intensively disturbed the spruce swampforests than the surrounding stands on mineral soil especially in the southern boreal area. The higher ages of the swampforest trees would support this assumption. As the epiphyte communities have quite low rates of population growth and competitive displacement (cf. Lawrey 1991) lowering the frequency and intensity of disturbance could increase species diversity. This hypothesis is consistent with the dynamic equilibrium model of diversity proposed by Huston (1994). Both the peat-layer and extensive cover of Sphagnum spp. on the ground-layer sustain moister and more even microclimate near the ground in the spruce swamp forests than in the adjacent mesic forests on mineral soil. This is favourable to several moisture demanding Evo 1 (southern boreal) 35- - fi4.n W . fz- Swamp Maqin 10-20m Location >SOm Swamp II Margin 10.20 m Loclltion > SO m Fig. 3. Epiphyte species richness per tree (mean + S.D.) in four different locations (1) in SPNW swampforest, 2) at swampforest margin, 3) in forest on mineral soil c. 10-20 m and 4) at least 50 m off the spruce-swamp margin) for each of the four sites. The means for lichens and bryophytes are shown separately. Differences between the locations were tested with ANOVA (N = 8) and pairwise multiple comparisons with Tukey's HSD test. The letters above the bars indicate the grouping of the locations at 50/0 ignificance level. Kuhmo 3 (rnlddleboreal) Kuhmo 4 (mkldlo boreal) species growing at basal trunks of trees, such as the lichens Mycoblustus spp. and Ochrolechiu undrogyna as well as most hepatics. The minor differences in the tree stand structure between the spruce swampforest and the surrounding forests are most probably unimportant for explaining the observed patterns in species composition and diversity. The spruce swampforests seem to be less important as refugia for epiphytes in the middle boreal sites. Due to the more humid macroclimate most moisture demanding species survive well also on trees on mineral soil, Cliostornwnlepromn being the only evident exception. Furthermore, the forest stands in the middle boreal sites are generally older and have a longer continuity than the southern boreal stands and even the stands on mineral soil have not been disturbed by forestry. The fire frequency and intensity may also have been lower in the humid forests of the middle boreal sites. Thus, most species should have had enough time to colonize trees in a wide area outside the swampforests due to the absence of severe disturbances. Kuusinen (1994a) also observed higher lichen species richness on Su1i.r cupreu in middle than in southern boreal Finland. The effect of air pollution may to some extent explain the lower lichen species richness in the southern boreal sites. The background sulphur deposition values in 1987 were c. 800 mg (S)m-* yr-' in the southern boreal ECOGRAPHY 1 9 1 11996) Fig. 4. Jackknife estimates of the Shannon-Wiener diversity index of (mean f S.D.) epiphyte communities in four diferent locations ( I ) in spruce swamp-forest, 2) at swampforest margin, 3) in forest on mineral soil c. 10-20 m and 4) at least 50 m off the spruce-swamp margin) for each of the four sites. DiRerences between the locations were tested with ANOVA (N = 8) and paimise multiple comparisons with Tukey’s HSD test. The letters above the points indicate the grouping of the locations at 5% significance level. Evo 1 (southem boreal) F4.99 p.lo.007 Evo 2 (southern boreal) F4.87 p4.009 Swamp ab I 8’ .SOm P[ Margin 10.20m Location Kuhmo 3 (middle boreal) Fd.96 pO.049 K u h m 4 (middle boreal) Swamp w.Iw Fa1.73 a [ Margin 1420m >50m Location and below 600 mg (S) m-2 yr-’ in the middle boreal sites (Tuovinen et al. 1990). The study sites were not affected by local air pollution sources. Especially lichens of the genera Alectoria, Bryoria and Usnea are known to be very sensitive to acid deposition (Hawksworth and Rose 1970, Insarova et al. 1992). Kuusinen et al. (1990) observed that these lichens have decreased markedly in southern Finland. Thus, the lower abundance of Alectoria sarmentosa, Bryoria spp. and Usnea spp. in the southern boreal sites may be due to the long-term effects of acid deposition. However, the minor difference in bark pH values between the southern and middle boreal sites do not indicate significant air pollution effects. Positive edge-effect, e.g. higher species diversity in ECOGRAPHY 1 9 1 (1996) ecotone than in adjacent habitats, is a commonly observed feature of animal and plant communities (Odum 1971, Hams 1988). However, my data does not indicate any significant trend of this kind, most probably because the change in species composition from the mesic forest to spruce swampforest was small and that the ecotone itself was too slight to attract any additional “edge species”. Positive edgeeffects could probably be observed in epiphyte communities of ecotones between more extreme habitats, such as riparian habitats o margins between agriculr tural land and forests. In addition, on a larger landscape scale the spruce swampforest itself could be considered as an ecotone habitat between forest and open mire. 32.5- E 2- r‘ 0 $1.5- Swamp Evo 1 H47.70 Evo 2 l i d 13 7 Fig. 5. Maximum height of @.OOo Laanactis abietina (mean S.D.) on spruce 21.5- 5 * 1- 1- 0.5- 0.50- 0- trunks in four different locations (I) in spruce swampforest, 2) at swampforest margin, 3) in forest on mineral soil c. 10-20 m and 4) at least 50 m off the spruce-swamp margin) in the two southern boreal sites (Evo I and 2). Differences between the locations were tested with the Kruskal-Wallis test (N = 8) and painvise multiple comparisons with Tukey’s HSD test. The letters above the bars indicate the grouping of the locations at 5% significance level. Margin 10-20rn Contrd Location Swamp Margin 10-20rn Control LocadiOn Species composition The crustose lichens Cliostomw leprosw, Arthonia leucopellaea and Lecanactis abietina were the most common species that were mainly confined to the spruce swamp-forests. The two latter species have a mainly southern boreal distribution in Finland, but C. leprosum is most abundant in the middle and northern boreal parts of the country. These species have been considered as indicatoxs of old-growth coniferous forests in Fennoscandia (Tmsberg 1992, Kuusinen 1994~. Skogsstyrclsen 1994). In southern boreal Finland C. leproswn and A. leucopellaea typically occur in small patches among an almost continuous cover of L. abietina. In northern parts of the country, where L. abietina is absent, C. leprosum forms alone large continuous patches on the basal trunks of especially the oldest spruces in swampforest stands. The sterile crustose lichens L.epraria jackii and Lecideo nylanderi replace these species on basal trunks of old spruces on mineral soil. The epiphyte species composition on the tms in the southern boreal spruce swampforests closely resembles the Lecanactidetum abietinae association characterized by Lecanactis abietina, Ochrolechia androgyna, Mycoblastus sanguinarius and Arthonia leucopellaea (Barkman 1958). According to Barkman (1958) this association occurs on bases of old Abies and Picea trunks in dense., moist, primeval forests in the upper montane zone in central Europe. According to Szlmermaa (1972) the above mentioned species are also typical on spruces in alder swamps of Estonia. Some rare epiphyte species, such as Evernia divaricata and Ramalina thrausta, which are typical in oldgrowth spruce swampforests may be very sensitive to any changes in the microclimate. According to Sjbberg and Ericson (1992) E. diuaricata rarely survived in areas where the surrounding forests were clear-cut even if the habitat itself was left intact. Thus, the conservation of surrounding forests may be most important in preserving epiphyte diversity in spruce swamp-forests. McCune (1993) has suggested that epiphyte species may be ordered similarly on three separate gradients: 1) vertical occurrence, 2) stand moisture and 3) stand age gradient. Two general trends occurring in my data are compatible with this “similar gradient hypothesis”. Firstly, the moisture favouring species, especially Lecanacris abietina, grew higher on the trunks on the swampforest t m s than on trees on mineral soil in the southern boreal sites. Secondly, most species that were mainly confined to swampforest trees in the southern boreal sites occurred also on forest trees in the more humid middle boreal sites. However, the vertical shift of xeric species upward on the trunks in the more humid stands as predicted by the hypothesis was not studied. Conclusions The,mults of the two southern boreal sites support the hypothesis that epiphyte species richness and diversity is higher in spruce swampforests than in adjacent mesic forests on mineral soil. This trend was, however, not ECOGRAPHY 1 9 1 (1996) observed in the two more humid and virgin middle boreal sites. Although any general pattern of species diversity and richness could not be detected the results suggest that old-growth spruce swamp-forests support a rich epiphyte flora. There seem to be several rare epiphyte species on spruce which may not survive outside these humid refugia, especially in the southern boreal Finland. Furthermore, Populus tremula and Sa1i.r caprea, two tree species with a diverse epiphyte flora (Kuusinen 1994a, b), often occur in or near spruce swamp-forests. These areas are thus one of the most important habitats for maintaining the epiphyte diversity in the boreal forest landscape. Vitikainen and K. Lohtander for their help in the identification of some difficult lichen specimens. K. Lohtander and R. Turpeinen are thanked for their help in the field and laboratory work. Thanks are also due to Lammi Biological Station and the Finnish National Board of Forestry for providing field equipment. study permissions and accommodation. P. -A. Essnn gave valuable comments on the manuscript. The research has been financed by the Academy of Finland and the Finnish Ministry of Environment. Acknowledgemenis - I am grateful to T. Tensberg, T. Ahti. 0. sensitivity and air pollution - a review of literature data. Swedish Environmental Protection Agency, Report 4007: 1-72. Kalela. A. 1961. Waldvegetationszonen Finnlands und ihrc klimatischen Paralleltypen. - Arch. Soc. 2001.Bot. Fennicae Vanamo 16 (Suppl.): 65-83. Koponen, T., Isoviita, P. and Lammes, T. 1977. The bryophytes of Finland: An annotated checklist. - Flora Fennica 6: 1-77.Koskinen, A. 1955. Uber die Kryptogamen der Biiume, besond n die Flechten. im Gewhergebiet des Piijlnne sowie an e den FlDssen Kalajoki. Lestijoki und Pyhfjoki. Floristische. soziologische und cikologische Studie I . - Memtor, Helsinki. Kuusinen. M. 1994a. Epiphytic lichen diversity on Salix caprea in old-growth s u h m and middle boreal forests of ote Finland. - Ann. Bot. Fennici 31: 77-92. 1994b. Epiphytic lichen flora and diversity on Populus rremula in old-growth and managed forests of southern and middle boreal Finland. - Ann. Bot. Fennici 31: 245260. 1994c. Metsiitalouden vaikutus epifyyttijikilllajiston mon- imuotoisuuteen (EHects of forestry on epiphytic lichen diversity). - Metsiintutkimuslaitoksen Tiedonantoja 482: 75-81. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Ecography Wiley

Importance of spruce swamp‐forests for epiphyte diversity and flora on Picea abies in southern and middle boreal Finland

Ecography , Volume 19 (1) – Mar 1, 1996

Loading next page...
 
/lp/wiley/importance-of-spruce-swamp-forests-for-epiphyte-diversity-and-flora-on-e0V0pq7405

References (32)

Publisher
Wiley
Copyright
Copyright © 1996 Wiley Subscription Services, Inc., A Wiley Company
ISSN
0906-7590
eISSN
1600-0587
DOI
10.1111/j.1600-0587.1996.tb00153.x
Publisher site
See Article on Publisher Site

Abstract

Kuusinen. M. 1996. Importance of spruce swamp-forests for epiphyte diversity and flora on Piceu ubies in southern and middle boreal Finland. - Ecography 1 9 41-51. The epiphyte (lichens and bryophytes) species richness, diversity and composition on basal trunks of Piceu ubies in spruce swamp-forests were compared to adjacent mesic forests on mineral soil in two southern and two middle boreal sites in old-growth forest patches in Finland. The sampling was carried out along four line transects parallel to swamp-forest margin: I ) in the spruce swampforest. 2) at the swamp forest margin, 3) on mineral soil c. 10-20 m OH the swampforest margin and 4) on mineral soil at least 50 m OHthe swampforest margin. In the two southern boreal sites there was a decreasing trend in the average species number per tree from the trees in swamp-forests (21-25 species) to the trees on mineral soil (17-18 species), whereas in the two more humid and virgin middle boreal sites a similar trend was not detected (25-28 species on all trees). There were no major diHerences in the epiphyte flora between the locations or study sites; the dominant species occurred on nearly all sample trees. The crustose lichens Cliusromum leprostun, Arrliuniu leuropellueu and Lecunuctis ubierinu were the most common species that were mainly confined to the swampforests in the southern boreal sites, C.leproswn also in the middle boreal sites. In addition. several rare species occurred exclusively on the swampforest trees. Lecunucris ubietinu extended significantly higher on the swampforest trees than on the trees on mineral soil in the southern boml sites. The spruce swampforests proved to be one of the most important habitats for maintaining the epiphyte diversity in the boreal forest landscape. hf. Kuusinen. Dept of Ecology und Systemutics. P.O. Box 47. Helsinki. Finland. FIN-OOO14 Uniu. oJ' Fire refugia areas with long continuity of old-growth forest are among the most important habitats in preserving natural species diversity in the boreal forest landscape of Fennoscandia (Esseen et al. 1992, Haila 1994),. Spruce swamp-forests have often been considered as such potential fire refugia (Zackrisson 1977, Zackrisson and Ostlund 1991, Sj6berg and Ericson 1992, Esseen et al. 1992, Angelstam and Rosenberg 1993, Hiirnberg et al. 1999, although some contrasting opinions have recently been presented (Segerstrbm et al. 1994, Oldhammer 1994). Hbrnberg et al. (1995), who studied the long-term continuity of ten old-growth spruce swamp-forest stands in northern Sweden, concluded that swamp-forests were less frequently affected Accepted I2 July 1995 Copyright 0 ECOGRAPHY 1996 ISSN 0906-7590 Printed in Ireland - all rights reserved ECOGRAPHY 1 9 1 (1996) by fire than surrounding forests on mineral soil. Some of their study areas had even been untouched by fire since the accumulation of peat began several thousand yean ago. The human impact was also slight in their study areas. Thus, the spruce swamp-forests may have exceptionally Long stand continuity in the boreal forest landscape and the probability for survival of species requiring long continuity should be higher in swamp forests than in adjacent mineral soil stands. Drainage has already destroyed a major part of Finnish spruce swampforests. Particularly small, thinpeated patches surrounded by mesic forests on mineral soil have been extensively drained. Eurola et a]. (1991) estimated that only c. 25% of the original spruce swamp-forest area (almost I million’ha) was still virgin in southern and central Finland. They considered several types of spruce swamp-forests to have become threatened or rare in Finland. Sjiiberg and Ericson (1992) have emphasized the importance of spruce swampforests and other forested wetlands for various organisms in boreal Fennoscandia. For example, a large number of moisture demanding cryptogams, both lichens and bryophytes, are confined to these habitats. In addition, several threatened species in Finland inhabit spruce swampforests (Rassi and Viisiinen 1987). Although several aspects of the floristic and faunistic composition of swampforests have been studied fairly well in Fennoscandia (Sjdberg and Ericson 1992). the epiphyte communities have received little attention. Only Koskinen (1955) and Somermaa (1972) have presented data on the epiphyte flora of swamp forests in their extensive studies in central Finland and Estonia. The aim of this study was to compare epiphyte species diversity, richness and composition between old Picea d i e s trees 1) in spruce swampforests, 2) at the margin of the swampforest and 3) in adjacent mesic forests on mineral soil. Two particular hypothesis of epiphyte species diversity and richness were tested: 1) whether diversity and richness are higher in the spruce swampforest than in the adjacent mesic forest on mineral soil, or 2) the ecotone between the spruce swamp forest and the mesic forest has the highest diversity and richness due to the presence of species characteristic of both habitats. Epiphyte species mainly confined to spruce swampforests o their margins were particularly r searched for. Study sites were selected in both southern and middle boreal Finland to detect geographical differences in species diversity, richness and composition. This study is part of a project searching for the keystone habitats for epiphyte diversity in the oldgrowth boreal forests of Finland. The spruce swamp-forest stands were small, c. I ha patches, in slight depressions surrounded by forests on mineral soil (Fig. 2). The peat layer of the spruce swampforests was generally thin ( e l m) and true forest species were common in the field layer. The ground layer was mostly covered by Sphagnum spp., especially S girgensohnii. Instead of the term “spruce . swampforest” defined by H6rnberg et al. (1995), the term “spruce mire” has often been used of these habitats in the Finnish literature (Eurola et al. 1984, 1991). The most common spruce swamp-forest types were the thin-peated spruce (KgK) and herb-rich forests (LhK) as well as Vaccinium myrtillus (MK) and Equisetum syluaticwn spruce mire (MkK) with some more herbrich patches (for the Finnish mire type classification and the type symbo1s;see Eurola et al. 1984). The spruce swampforests of the two southern boreal sites Study sites Four study sites were selected in old spruce-dominated forest areas, two (1, 2) in the southern boreal and two (3, 4) in the middle boreal zone (Fig. 1). The southern boreal sites Evo 1 and 2 were located c. 1 km apart from each other in a c. 300 ha old forest patch and the middle boreal sites Kuhmo 3 and 4, c. 4 lun apart from each other in c. 100 ha and c. 80 ha patches of old-growth forest. Managed forests, clear-cuts and fires surrounded the forest patches. In the southern boreal sites some human impact, such as signs of old selective logging, were visible, but the two middle boreal sites were of primeval nature. Each study site included part of a spruce swampforest stand and its margin as well as part of adjacent mesic forest stand on mineral soil. The total area of the sites was 1.5-3 ha. Fig. 1. The location of study sites. The thick lines show the borders between vegetation zones (Ahti et al. 1968): A = hemiboreal. B = southern boreal, C = middle boreal, D = northern boreal. ECOGRAPHY 1 9 1 (19961 Fig. 2. General view of ecotone between spruce swampforest and mesic forest on mineral soil close by the southern boreal site Evo 1. were slightly drained in the late 19th century. The dominant forest site types on the mineral soil were the herb-rich Oxalis-Myrtillus (OMT) and the mesic M yrtillus type (MT) in the southern boreal and the corresponding Geranium-0xalis-Myrtillus (GOMT) and Vuccinium-Myrtillus type (VMT) in the middle boreal sites (for the Finnish forest site type classification, see Kalela 1961). The altitude of the two southern boreal sites was c. 160 m a.s.1. and the middle boreal sites c. 260 m a d . The maximum difference in altitude within each study site was between 5 and I5 m. The mean annual temperature is c. +4"C in the southern boreal sites and c. 1°C in the middle boreal sites (Alalarnmi 1987). The coldest months are January and February with mean temperatures of c. -8°C in the southern and c. - 11°C in the middle boreal sites; the warmest month is July with mean temperatures of c. 17°C and c. 1516°C. respectively. The mean duration of the growing season (number of days with mean temperature exceeding +5"C) is c. 165 d in the southern boreal sites and c. 140 d in the middle boreal sites. The mean annual precipitation is 650-700 rnm in all sites, the wettest months being July and August and the driest ones February and March. The mean maximum snow depth in forests is c. 50 cm in the southern and c. 80 cm in the middle boreal sites. The most significant climatic difference between the southern and middle boreal sites is the higher humidity, measured as the difference between precipitation and evaporation, in the latter sites. This value, counted from the disappearance of snow in spring to the end of July, is c. -60 mm in the southern and c. +40 rnm in the middle boreal sites (Alalammi 1987). Methods Sampling The field work was carried out in the summers of 1993 and 1994. The sampling was performed along four 100-200 m long line transects parallel with the spruce swampforest margin: 1) in the spruce swampforest c. 20 m off the swampforest margin. 2) at the swamp f forest margin, 3) c. 10-20 m o f the swampforest margin on mineral soil and 4) at least 50 rn and up to 200 m o f the swampforest margin on mineral soil. f Eight spruces closest to random points were sampled along each of the four transects. Only straight, living trees with a diameter at breast height (DBH) of at least 18 cm were sampled. Trees with plenty of twigs below two meters height on the trunk were also avoided. The Occurrence of all lichen and bryophyte species on the basal trunk < 2 m in height was recorded on the sample trees. The structure of the epiphyte f o a was lr surveyed more in detail by placing a rectangle made of transparent plastic on each tree above 80 cm in height, on both the south and north sides of the trunk. The 20 x 50 cm rectangle was divided into 100 2 x 5 cm subunits, and a single point was randomly placed in each of these. The number of points hit by each species were counted and these values were used as an estimate of the species' cover percentage on the sample plot. The maximum height of the continuous cover of the CNStose lichen Lecanactis abietina was also measured on each sample tree. Specimens of difficult crustose species were collected for identification with microscope and standard thin layer chromatography (White and James 1985). ECOGRAPHY 1 9 1 (1996) Voucher specimens were deposited at the Botanical Museum of Univ. of Helsinki (H). DBH and height of each sample tree were measured. The canopy cover was estimated in lO%-classes adjacent to each sample tree. The distance from the sample tree to the three closest living trees (DBH > 4 cm) was measured to obtain an estimate of stem density. Age of two or three sample trees on each of the four transects and in each site were measured with an increment borer. Bark samples were collected from the trunks at 0.5-1.5 m above ground outside the sample plots. Three grams of bark was soaked in 37 ml of distilled water for 24 h and the pH value of the solution was measured with a standard pH meter. Data analysis Species diversity at each location was calculated with the Shannon-Wiener diversity index (H’). The cover estimates on the south and north sides of the trunk were combined for each species. The index was calculated using the formula swampforest stands and forest stands on mineral soil were in tree age and stem density (Table I). Age showed a decreasing gradient from the swamp-forest to the t m s on mineral soil in both the southern boreal sites and the middle boreal site 4. The swampforest stands tended to be slightly more open than the forest stands on mineral soil. This was reflected by the lower crown cover and higher distance between t m s in the swamp-forest stands of sites 2, 3 and 4 (Table 1). The sample trees were thinner and shorter in the middle boreal than in the southern boreal sites: the mean DBH was 32 cm and tree height 27 m in the southern boreal sites but only 25 cm and 19 m in the northern boreal sites (Table I). However, the middle boreal trces were generally older than the southern boreal ones: the oldest middle boreal swamp forest tms in the data approached 300 yr while the youngest southern boreal ones were < 100 yr old. The acidity of the bark was slightly >pH 3 on most of the sample trees. The bark pH of the sample trees on site 1 was on an average somewhat lower than on the other sites. H I =i- I pi Inp, Species richness and diversity where pi is the proportion of ith species of total cover. The jackknife procedure proposed by Zahl (1977) and Routledge (1980) was used to estimate the diversity indices and to attach confidence limits to these estimates. Statistical comparisons between the four locations were performed with either the parametric one-way analyses of variance (ANOVA) or the non-parametric Kruskal-Wallis test. Pairwise multiple comparisons between the locations were performed with Tukey’s HSD test using 5% significande level. The significance of the differencesin the occurrence of individual species were tested with the Fisher’s exact test for 2 x 2 tables. Trees on the spruce swampforest and at the swampforest margin were compared to all the trees on mineral soil. The presence and absence of each species formed the other dimension of the table. Data from the two southem boreal and two middle boreal plots were combined in this analyses to accomplish larger sample sizes. All the analyses were carried out with Systat for Windows (Wilkinson et al. 1992). The nomenclature follows Santesson (1993) for lichens and Koponen et al. (1977) for bryophytes. A total of 48 lichen and 12 bryophyte species were found on the 64 southern boreal trees and 60 lichen and 12 bryophyte species on the 64 middle boreal trees (Table 2). In addition, two conspicuous non-lichenized fungi, Arthothelium scandinavicum and Chaenothecopsis viridiulba, were included in the data. In the two southern boreal sites the mean species number per tree was significantly higher on the swampforest trees than on the t m s on mineral soil, whereas a similar trend could not be detected in the middle boreal sites (Fig. 3). The mean species numbers per tree wen generally higher on the middle boreal than on the southern boreal trees: the mean species number was 27 on the middle boreal but only 19 on the southern boreal t m . This difference was largest on the trees on mineral soil. The total number of species also showed a decreasing trend from the swampforests to the forests on mineral soil especially in the southern boreal sites (Table 2). The observed differences were mainly in the lichen flora, whereas the bryophyte species numbers varied less (Fig. 3, Table 2). The jackknife estimates of the Shannon-Wiener diversity index were also somewhat higher on the swampforest trees than on the trees on mineral soil in both southern boreal sites and the ResnltS middle boreal site 3 (Fig. 4). The large variation in the point frequency data, however, complicates the Sample tree and stand characteristics detection of any significant pattern in species diverThe most significant differences between the spruce sity. ECOGRAPHY 19:l (1996) Table I . Characteristics of the sampled Piceu uhies trees in each of the four study sites. Locations of the trees are from the left to right column: 1 ) spruce swamp-forest, 2) swamp-forest margin, 3) mineral soil c. 10-20 m OH the swamp-forest margin and 4) mineral soil at least 50 m OH the swamp-forest margin. Comparisons between locations: I = ANOVA, = Knrskall-Wallis test. The last column shows probabilities for no difference between the locations. N = 8 for each measurement. except for tree age. Swamp EVO I (southern boreal) DBH (cm), mean f S.D. Tree height (m), mean f S.D. Acidity of bark (pH), mean f S.D. Mean distance to 3 closest trees (m). mean f S.D. Canopy cover (%), median Tree age (yr), range (N) EVO 2 (southern boreal) DBH (cm). mean f S.D. Tree height (m). mean f S.D. Acidity of bark (pH), mean f S.D. Mean distance to 3 closest trees (m), mean f S.D. Canopy cover (%), median Tree age (yr). range (N) KUHMO 3 (middle boreal) DBH (cm). mean f S.D. Tree height (m), mean f S.D. Acidity of bark (pH), mean f S.D. Mean distance to 3 closest trees (m). mean f S.D. Canopy cover (YO). median Tree age (yr), range (N) KUHMO 4 (middle boreal) DBH (cm). mean f S.D. Tree height (m). mean f S.D. Acidity of bark (pH), mean f S.D. Medn distance to 3 closest trees (m). mean fS.D. Canopy cover (YO). median Tree age (yr), range (N) 32 f 5 26 f 2 3.1 f0.1 3.0 f0.7 80 145-164 ( 3 ) 35f6 28 f 3 3.3 f0.2 4.2 f0.7 65 145-158 (3) 27 f 5 2Of2 3.4 f0. I 3.3f 1 . 1 60 139-175 (3) 29 f 7 19f2 3.520.1 3.3 f 0.7 50 200-282 (2) Margin 28f6 24 f2 3.1 f 0.2 3.1 f 1.1 75 126-160 (2) 34f7 28 f 2 3.3 f 0 . l 3.9 f 0.9 70 130- 148 (2) 24f4 19f I 3.5f0.1 3.4 f0.8 55 143-172 (3) 26 k 6 19f3 3.3 f 0.2 2.5 f0.6 60 185-200 (2) 10-20 m 50 m off the swampforest margin) are marked with M in the pattern column. The significance of these patterns were tested with the Fisher's exact test for 2 x 2 tables. Significance values: three symbols = p < 0.001. two symbols = p < 0.01, one symbol = p < 0.05; N = 32. Bryophytes are marked with an asterisk. Species Hypogymnia physodes Lepraria spp. Parmeliopsis ambigua Cladonia coniocraea etc. Cladonia cenotea Panneliopsis hyperopia Chaenatheca chrysocephala Pleuro:ium schreberis Pilidiwn pulcherrimwn8 Clahnia digiiaia Plagiotheciwn Iaeium' Platismaiia glauca Lmospora elatina Ochrolechia microsiicioides Bryoria capillarb M.vcobIasr~rsanguinarius Vulpicida pinastri Bryoria fuseucens Dicranwn scopariwn' i+lycoblastus ofinis Dicranwn furccsecns8 Ochrolechia an&ogyna Lecidea nylanderi Micarea prasina Usnea spp. Arthothelium scandinavicum Cladoniafimbriaia Chaenotheca subroscida Lecanuctis abieiina Cliostomum leprosum Brachyiheciwn spp.+ Alectoria sanneniosa Caliciwn uiride Lecidea albofuscescens Ceiraria chlorophylla L0pho:ia longidens' Lecanora hypopta coll. Micarea melaena Caliciwn glaucellwn Arthonia leucopelhea k i a k a Iepranoides Chaenotheca irichialis Pertusaria boreal& Mycoblarm alpinus Arthonia incanrota Biaiom efirescens Japewia subaurvera Lecanora circumborealis Biaiora gyrophorica Cypheliwn inquinans Orihodicramun monimum* Dimerella pineii Hylocomium splendens* Ochrolechia albcflawscem Pohlia nutans' Chaenotheca sternonea Hypocenomyce scalaris Hypogyrnnia tubulosa Lophocolea heierophylla' Pertusaria pupillaris Bryoria furcellata Chaenothecaferruginea Lepidoria reptans' Micarea denigraia Swamp Margin Evo 1-2 (southern boreal) Kuhmo 3-4 (middle boreal) 10-20 m >50 m Pattern Swamp Margin 10-20 m >M m Pattern 1 6 16 1 6 1 6 16 I5 16 13 13 16 1 6 12 12 IS 16 I5 II 1 4 1 6 16 16 16 14 16 13 16 14 I2 9 16 16 16 1 6 1 6 16 I5 I5 14 IS 1 6 I5 1 6 IS 16 sss MM I5 6 I5 II 7 3 sss ss 16 I5 13 1 0 14 14 7 13 8 1 6 I2 14 7 6 9 14 9 MM MM MM MM a a sss a a 6 II 5 7 MMM ss 2 2 I 3 Table 2. (cont.) Species Swamp Margin 10-20 m Evo 1-2 (southem boreal) >50 m Pattern Swamp Margin Kuhmo 3-4 (middle boreal) 10-20 m >50 m Pattern I Varicellaria rhodocarpa Bryoria jremontii Chaenotheca furfuracea Imshaugia aleurites Parmelia sulcata Pertusaria amara Pertusaria coccodes Pseudevernia furfuracea Ramalina thrausta Tetraphu pellucida' Arthonia didyma Barbilophoria attenuata' Caliciwn pannun Chaenotheca Iaevigata Charnothecopsis viridialba Cladonia cornuta Cladonia pleurota Lecanora expallens Lecidea erythrophaea Lopadiwn disciforme Sanionia uncinata' Number of lichens Number of bryophytes Total number of species I I cantly higher on the swampforest trees than on the trees on mineral soil in the two southern boreal sites (Fig. 5). In swamp-forests this species may completely veil the basal trunks of spruces up to 3 m height while it is often absent on spruces on mineral soil. This southern boreal species did not occur in the middle boreal sites. Species richness and diversity The numbers of lichen species observed in this study were relatively high for Picea d i e s . especially in the middle boreal sites (60species on 64 trees). For example, HyvHrinen et al. (1992) found only 38 lichen species on over 100 spruces and Sbmermaa (1972) reported 74 lichen species on spruce in her extensive data for hemiboreal Estonian forests. Koskinen (1955) reported a total of 154 lichen species on Picea d i e s in his study area in central Finland. This very high species number is, however, not quite comparable with my data. His data had been collected from a very large number of trees (though the figures are not given) in various habitats and includes species on both the whole trunk and branches. Furthermore, several of his species are of doubtful taxonomic status, such as many Biatora spp. and Umea spp. Kuusinen (1994a, b) studied the lichen species numbers per tree on Salix caprea and Populus tremula using the same methodology and partly the same areas as in this study. The lichen species numbers per tree were generally higher on Salk caprea (mean 22) and lower on Populus tremula (mean 13) than on Picea abies. ECOGRAPHY 1 9 1 (1996) There are only few estimates of the epiphyte species richness in wet boreal forests. According to Somermaa (1972) alder swamps had the highest epiphytic lichen richness of all Estonian forest types. Koskinen (1955) did not separate thin-peated spruce swampforests of mesic forests in his data, but wet spruce swampforests and pine bogs werc the most epiphyte species poor habitats in his data. He concluded that the low species numbers were mainly due to the scarcity of trees, high light intensity and drought. These habitats are thus quite different from my study sites which had a dense tree stand and closed canopy. Two prominent factors behind the observed high epiphyte species richness and diversity in the spruce swampforests could be: 1) the long continuity and 2) the humid microclimate of the swampforest stands. Forestry and fires, probably also storms, may have less frequently and intensively disturbed the spruce swampforests than the surrounding stands on mineral soil especially in the southern boreal area. The higher ages of the swampforest trees would support this assumption. As the epiphyte communities have quite low rates of population growth and competitive displacement (cf. Lawrey 1991) lowering the frequency and intensity of disturbance could increase species diversity. This hypothesis is consistent with the dynamic equilibrium model of diversity proposed by Huston (1994). Both the peat-layer and extensive cover of Sphagnum spp. on the ground-layer sustain moister and more even microclimate near the ground in the spruce swamp forests than in the adjacent mesic forests on mineral soil. This is favourable to several moisture demanding Evo 1 (southern boreal) 35- - fi4.n W . fz- Swamp Maqin 10-20m Location >SOm Swamp II Margin 10.20 m Loclltion > SO m Fig. 3. Epiphyte species richness per tree (mean + S.D.) in four different locations (1) in SPNW swampforest, 2) at swampforest margin, 3) in forest on mineral soil c. 10-20 m and 4) at least 50 m off the spruce-swamp margin) for each of the four sites. The means for lichens and bryophytes are shown separately. Differences between the locations were tested with ANOVA (N = 8) and pairwise multiple comparisons with Tukey's HSD test. The letters above the bars indicate the grouping of the locations at 50/0 ignificance level. Kuhmo 3 (rnlddleboreal) Kuhmo 4 (mkldlo boreal) species growing at basal trunks of trees, such as the lichens Mycoblustus spp. and Ochrolechiu undrogyna as well as most hepatics. The minor differences in the tree stand structure between the spruce swampforest and the surrounding forests are most probably unimportant for explaining the observed patterns in species composition and diversity. The spruce swampforests seem to be less important as refugia for epiphytes in the middle boreal sites. Due to the more humid macroclimate most moisture demanding species survive well also on trees on mineral soil, Cliostornwnlepromn being the only evident exception. Furthermore, the forest stands in the middle boreal sites are generally older and have a longer continuity than the southern boreal stands and even the stands on mineral soil have not been disturbed by forestry. The fire frequency and intensity may also have been lower in the humid forests of the middle boreal sites. Thus, most species should have had enough time to colonize trees in a wide area outside the swampforests due to the absence of severe disturbances. Kuusinen (1994a) also observed higher lichen species richness on Su1i.r cupreu in middle than in southern boreal Finland. The effect of air pollution may to some extent explain the lower lichen species richness in the southern boreal sites. The background sulphur deposition values in 1987 were c. 800 mg (S)m-* yr-' in the southern boreal ECOGRAPHY 1 9 1 11996) Fig. 4. Jackknife estimates of the Shannon-Wiener diversity index of (mean f S.D.) epiphyte communities in four diferent locations ( I ) in spruce swamp-forest, 2) at swampforest margin, 3) in forest on mineral soil c. 10-20 m and 4) at least 50 m off the spruce-swamp margin) for each of the four sites. DiRerences between the locations were tested with ANOVA (N = 8) and paimise multiple comparisons with Tukey’s HSD test. The letters above the points indicate the grouping of the locations at 5% significance level. Evo 1 (southem boreal) F4.99 p.lo.007 Evo 2 (southern boreal) F4.87 p4.009 Swamp ab I 8’ .SOm P[ Margin 10.20m Location Kuhmo 3 (middle boreal) Fd.96 pO.049 K u h m 4 (middle boreal) Swamp w.Iw Fa1.73 a [ Margin 1420m >50m Location and below 600 mg (S) m-2 yr-’ in the middle boreal sites (Tuovinen et al. 1990). The study sites were not affected by local air pollution sources. Especially lichens of the genera Alectoria, Bryoria and Usnea are known to be very sensitive to acid deposition (Hawksworth and Rose 1970, Insarova et al. 1992). Kuusinen et al. (1990) observed that these lichens have decreased markedly in southern Finland. Thus, the lower abundance of Alectoria sarmentosa, Bryoria spp. and Usnea spp. in the southern boreal sites may be due to the long-term effects of acid deposition. However, the minor difference in bark pH values between the southern and middle boreal sites do not indicate significant air pollution effects. Positive edge-effect, e.g. higher species diversity in ECOGRAPHY 1 9 1 (1996) ecotone than in adjacent habitats, is a commonly observed feature of animal and plant communities (Odum 1971, Hams 1988). However, my data does not indicate any significant trend of this kind, most probably because the change in species composition from the mesic forest to spruce swampforest was small and that the ecotone itself was too slight to attract any additional “edge species”. Positive edgeeffects could probably be observed in epiphyte communities of ecotones between more extreme habitats, such as riparian habitats o margins between agriculr tural land and forests. In addition, on a larger landscape scale the spruce swampforest itself could be considered as an ecotone habitat between forest and open mire. 32.5- E 2- r‘ 0 $1.5- Swamp Evo 1 H47.70 Evo 2 l i d 13 7 Fig. 5. Maximum height of @.OOo Laanactis abietina (mean S.D.) on spruce 21.5- 5 * 1- 1- 0.5- 0.50- 0- trunks in four different locations (I) in spruce swampforest, 2) at swampforest margin, 3) in forest on mineral soil c. 10-20 m and 4) at least 50 m off the spruce-swamp margin) in the two southern boreal sites (Evo I and 2). Differences between the locations were tested with the Kruskal-Wallis test (N = 8) and painvise multiple comparisons with Tukey’s HSD test. The letters above the bars indicate the grouping of the locations at 5% significance level. Margin 10-20rn Contrd Location Swamp Margin 10-20rn Control LocadiOn Species composition The crustose lichens Cliostomw leprosw, Arthonia leucopellaea and Lecanactis abietina were the most common species that were mainly confined to the spruce swamp-forests. The two latter species have a mainly southern boreal distribution in Finland, but C. leprosum is most abundant in the middle and northern boreal parts of the country. These species have been considered as indicatoxs of old-growth coniferous forests in Fennoscandia (Tmsberg 1992, Kuusinen 1994~. Skogsstyrclsen 1994). In southern boreal Finland C. leproswn and A. leucopellaea typically occur in small patches among an almost continuous cover of L. abietina. In northern parts of the country, where L. abietina is absent, C. leprosum forms alone large continuous patches on the basal trunks of especially the oldest spruces in swampforest stands. The sterile crustose lichens L.epraria jackii and Lecideo nylanderi replace these species on basal trunks of old spruces on mineral soil. The epiphyte species composition on the tms in the southern boreal spruce swampforests closely resembles the Lecanactidetum abietinae association characterized by Lecanactis abietina, Ochrolechia androgyna, Mycoblastus sanguinarius and Arthonia leucopellaea (Barkman 1958). According to Barkman (1958) this association occurs on bases of old Abies and Picea trunks in dense., moist, primeval forests in the upper montane zone in central Europe. According to Szlmermaa (1972) the above mentioned species are also typical on spruces in alder swamps of Estonia. Some rare epiphyte species, such as Evernia divaricata and Ramalina thrausta, which are typical in oldgrowth spruce swampforests may be very sensitive to any changes in the microclimate. According to Sjbberg and Ericson (1992) E. diuaricata rarely survived in areas where the surrounding forests were clear-cut even if the habitat itself was left intact. Thus, the conservation of surrounding forests may be most important in preserving epiphyte diversity in spruce swamp-forests. McCune (1993) has suggested that epiphyte species may be ordered similarly on three separate gradients: 1) vertical occurrence, 2) stand moisture and 3) stand age gradient. Two general trends occurring in my data are compatible with this “similar gradient hypothesis”. Firstly, the moisture favouring species, especially Lecanacris abietina, grew higher on the trunks on the swampforest t m s than on trees on mineral soil in the southern boreal sites. Secondly, most species that were mainly confined to swampforest trees in the southern boreal sites occurred also on forest trees in the more humid middle boreal sites. However, the vertical shift of xeric species upward on the trunks in the more humid stands as predicted by the hypothesis was not studied. Conclusions The,mults of the two southern boreal sites support the hypothesis that epiphyte species richness and diversity is higher in spruce swampforests than in adjacent mesic forests on mineral soil. This trend was, however, not ECOGRAPHY 1 9 1 (1996) observed in the two more humid and virgin middle boreal sites. Although any general pattern of species diversity and richness could not be detected the results suggest that old-growth spruce swamp-forests support a rich epiphyte flora. There seem to be several rare epiphyte species on spruce which may not survive outside these humid refugia, especially in the southern boreal Finland. Furthermore, Populus tremula and Sa1i.r caprea, two tree species with a diverse epiphyte flora (Kuusinen 1994a, b), often occur in or near spruce swamp-forests. These areas are thus one of the most important habitats for maintaining the epiphyte diversity in the boreal forest landscape. Vitikainen and K. Lohtander for their help in the identification of some difficult lichen specimens. K. Lohtander and R. Turpeinen are thanked for their help in the field and laboratory work. Thanks are also due to Lammi Biological Station and the Finnish National Board of Forestry for providing field equipment. study permissions and accommodation. P. -A. Essnn gave valuable comments on the manuscript. The research has been financed by the Academy of Finland and the Finnish Ministry of Environment. Acknowledgemenis - I am grateful to T. Tensberg, T. Ahti. 0. sensitivity and air pollution - a review of literature data. Swedish Environmental Protection Agency, Report 4007: 1-72. Kalela. A. 1961. Waldvegetationszonen Finnlands und ihrc klimatischen Paralleltypen. - Arch. Soc. 2001.Bot. Fennicae Vanamo 16 (Suppl.): 65-83. Koponen, T., Isoviita, P. and Lammes, T. 1977. The bryophytes of Finland: An annotated checklist. - Flora Fennica 6: 1-77.Koskinen, A. 1955. Uber die Kryptogamen der Biiume, besond n die Flechten. im Gewhergebiet des Piijlnne sowie an e den FlDssen Kalajoki. Lestijoki und Pyhfjoki. Floristische. soziologische und cikologische Studie I . - Memtor, Helsinki. Kuusinen. M. 1994a. Epiphytic lichen diversity on Salix caprea in old-growth s u h m and middle boreal forests of ote Finland. - Ann. Bot. Fennici 31: 77-92. 1994b. Epiphytic lichen flora and diversity on Populus rremula in old-growth and managed forests of southern and middle boreal Finland. - Ann. Bot. Fennici 31: 245260. 1994c. Metsiitalouden vaikutus epifyyttijikilllajiston mon- imuotoisuuteen (EHects of forestry on epiphytic lichen diversity). - Metsiintutkimuslaitoksen Tiedonantoja 482: 75-81.

Journal

EcographyWiley

Published: Mar 1, 1996

There are no references for this article.