- Email: [email protected]
Population responses of oribatid mites and collembolans after drought
Applied Soil Ecology 28 (2005) 163–174 www.elsevier.com/locate/apsoil
Population responses of oribatid mites and collembolans after drought N. Lindberg*, J. Bengtsson Department of Ecology and Crop Production Science, Swedish University of Agricultural Sciences, P.O. Box 7043, SE-750 07 Uppsala, Sweden Received 7 April 2004; received in revised form 4 July 2004; accepted 9 July 2004
Abstract To compare the effects of a drought disturbance on species of Oribatida and Collembola, and subsequent recovery of their populations after the drought, we examined a Norway spruce, Picea abies, stand in south-western Sweden, where 6 years of experimentally induced summer droughts had resulted in major changes in the soil faunal communities. We followed the population densities during a 4-year period and sought correlations between the species’ drought responses and their ecological characteristics. Data on depth preference, habitat choice and reproductive mode were collected from the literature. Surfaceliving species, which tended to have narrow habitat width, were less negatively affected by the drought. However, among species showing negative population responses to drought, species with large habitat widths tended to recover faster after the drought. Furthermore, parthenogenesis was more common among the oribatid species that showed a population recovery than among those that did not. Overall, collembolan species recovered faster than oribatids, and among the species that did not recover, Oribatida were over-represented. No general differences in characteristics between oribatids and collembolans were observed that could explain their different responses. Possibly, traits other than those examined were more important, such as differences in dispersal rates between the two groups. # 2004 Elsevier B.V. All rights reserved. Keywords: Collembola; Colonisation; Disturbance; Life history; Oribatida; Population recovery
1. Introduction Many ecosystems are subject to disturbances of varying frequencies, such as fire, drought, flooding and, when under human management, events like * Corresponding author. Tel.: +46 18 67 24 20; fax: +46 18 67 28 90. E-mail address: [email protected] (N. Lindberg).
clear-cutting or tilling. To persist in such ecosystems, organisms either have to survive the disturbance, or recolonise disturbed patches from the surroundings. Hence, the traits that allow species to survive in dynamic landscapes are likely to be different from those allowing species to persist locally (Loreau et al., 2003) and may be related to the colonising ability of the species. In colonisation theory, life-history traits have often been used to explain colonisation patterns
0929-1393/$ – see front matter # 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.apsoil.2004.07.003
164
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
in plants and animals (Ebenhard, 1991). Good colonisers, sometimes equated with ‘‘r-selected species’’, are often characterised by early age at first reproduction, short life span, large clutches and parthenogenesis, whereas the opposite has been suggested to be common among late colonisers, ‘‘K-selected species’’ (see, for instance, Baker, 1955; MacArthur and Wilson, 1967; Boyce, 1984; Grubb, 1987). However, the pattern is far from clear-cut, as landscape dynamics or traits such as dispersal ability and niche specialisation may also influence colonisation processes (Diamond, 1974; Simberloff, 1981; ˚ s et al., Baur and Bengtsson, 1987; Ebenhard, 1991; A 1992; Ronce et al., 2000). Post-disturbance colonisation and succession of soil fauna have received considerable interest over the years, and studies have focused, among others, on man-made habitats and the impact of pesticides, pollution and fire (e.g., Davis, 1986; Bengtsson and Rundgren, 1988; Hoy, 1990; Webb, 1994; Skubala, 1995; Wanner and Dunger, 2002). In soil microarthropods, dispersal behaviour, lifespan and reproductive strategies are probably key factors regulating survival and colonisation after disturbances (Norton, 1994; Petersen, 1995; Siepel, 1995; Shaw, 1997; St. John et al., 2002). It has also been suggested that the ability to survive in sheltered microsites is important (Tamm, 1984; Hopkin, 1997; Shaw, 1997). However, there have been few serious attempts to link the impact of disturbance and subsequent succession to the lifehistory traits and other characteristics of the species involved, probably because of a lack of data on many species (but see Siepel, 1996; Maraun and Scheu, 2000). Thus, general conclusions have often been drawn without proper tests. Oribatid mites and collembolans share many similar features as they are mainly fungivorous, have similar body sizes, and are very numerous groups in the organic soil layers. However, the general lifehistory traits of Oribatida have been considered typical of K-selected species (Norton, 1994), whereas collembolan species exhibit wider variation in lifehistory traits (Norton, 1994; Hopkin, 1997). Therefore, the species-wise responses of Oribatida and Collembola after a disturbance probably differ. It has been argued that disturbances are the main structuring force for oribatid mite communities (Acari: Oribatida) (Maraun and Scheu, 2000). Also,
effects of disturbances on soil faunal communities are often stronger and more long-lived in oribatids than in collembolans (Strojan, 1978; Hutson, 1980; Lucarotti, 1981; Lindberg et al., 2002; Lindberg and Bengtsson, unpublished). This may be a result of general differences in life-history, or differences in dispersal ability and habitat specialisation between the groups. To study the effects of a drought disturbance and the subsequent recovery of Oribatida and Collembola, we examined a forest site in south-western Sweden where experimentally induced long-term summer droughts had caused major changes in the soil faunal community (Lindberg et al., 2002). One advantage of the site was that the species pool consisted of many relatively well-known microarthropods. Our aim was to examine the extent to which a few life history and ecological characteristics could explain variations in the impact of drought on the soil fauna, and their recovery patterns after the disturbance. The main questions we addressed were: (1) Will the populations of collembolan species recover faster than oribatid species, as they are considered less K-selected (Norton, 1994)? (2) Will deep-living species recover faster than species living closer to the surface, as they are more likely to have survived the disturbance due to their ability to penetrate deeper into the soil during drought? (3) Will populations of parthenogenetic species recover faster than sexually reproducing ones, as suggested by r-selection theory? (4) Will generalist species recover faster than species with narrower habitat preferences, as suggested by ˚ s et al. (1992) for fragmented habitats in e.g., A general?
2. Material and methods 2.1. Site description The field study was carried out during 1996–1999 in a homogeneous Norway spruce (Picea abies) stand situated at Skogaby, south-western Sweden (568330 N, 138130 E), 16 km from the coast at an altitude of 95– 115 m above sea level. The climate is maritime with a
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
mean annual precipitation of about 1150 mm, and an annual mean temperature of 7.5 8C. The stand was a second-generation forest on former heathland. It was 32 years old when the present study started and the trees were planted as 2-year-old seedlings in 1966. For more details on soil conditions and stand history, see Lindberg et al. (2002). The stand was very dense, and due to the closed canopy less than 2% of the incident light penetrated to the ground. No field or shrub layer existed and the under-storey consisted of sparse mosses with scattered tussocks of grass. No changes in vegetation cover on the plots were detected during the studied period. The soil conditions and stand type at Skogaby are typical for large areas of managed forest in southern Sweden (Bergholm et al., 1995). 2.2. Experimental design We used plots that were part of a large field experiment established in 1988 to study effects of water and nutrient availability on tree growth (Nilsson, 1997). It had a randomised block design with four blocks, each containing one replicate of each treatment. The present study examined the effects of two of the seven treatments, namely control (no treatment) and summer drought (created by roofs covering most of the plot area). Plot size was 45 m 45 m for the control plots and, initially (1990–1996) 45 m 22.5 m for the drought plots. A recovery treatment was then initiated by permanently removing roofs from half of each drought plot’s surface, giving drought plots and recovery plots, both measuring 22.5 m 22.5 m. The drought treatment started in 1990 and the recovery treatment started in autumn 1996. The plastic roofs were transparent and placed 1–1.5 m above ground. They had openings for tree trunks and for maintenance purposes, but prevented 70% of the throughfall on the plots from reaching the ground during April to September (Alavi, 1999). During the winter the roofs were removed and all throughfall could reach the ground (Table 1), resulting in a summer drought treatment (see Lindberg et al., 2002 for more details). Litter falling on the roofs was spread on the ground when the roofs were removed in October. During the period of roof coverage approximately 40% of the yearly litterfall occurred (L-O. Nilsson, pers. comm.).
165
Table 1 Duration of the summer drought treatment and data on timing and sampled soil surface during the soil faunal samplings between 1996 and 1999 Year
Roofs on
Roofs off
Sampling
Sampled surface/plot
1996 1997 1998 1999
Mid-April Mid-April Mid-July Early June
Early October Early October Mid-December Early October
Oct. 29 Oct. 20 Nov. 12 Nov. 4
3 dm2 4 dm2 4 dm2 3 dm2
2.3. Sampling and species identification The plots were sampled once every year, in late autumn 1996–1999, 14–34 days after roof removal (Table 1). Arthropods were collected by taking three or four systematically distributed 100-cm2 10-cmdeep subsamples of the litter and humus layers from each plot, and the layers were separated. The sampling points were evenly distributed within the plots, placed 7.5 m from the edge on the lines between the centre of each side and the mid-point of the plot. When sampling in drought plots we avoided areas closer than 1 m to tree trunks and gaps in the roof coverage. In the laboratory the subsamples were stored at 4 8C for 12 h before extraction in Tullgren dry funnels for 4 days. Due to practical constraints all subsamples from each plot and layer were pooled before extraction in 1996– 1997, but this was not done in 1998–1999. The samples from 1998 were later excluded from the present study as the late start of the drought treatment in this year probably affected the drought reference and complicated recovery analysis (see Lindberg and Bengtsson, unpublished). All adult individuals of Collembola and Oribatida were determined to the species level (see Lindberg et al., 2002). Oribatid nomenclature follows Niemi et al. (1997), while collembolan nomenclature follows Fjellberg (1998) for Poduromorpha, and Fjellberg (1980) for other groups. All data can be found in Lindberg (2003). 2.4. Life-history and ecological traits To examine how species recovery was related to life-history and ecological characteristics, the following traits were related to the effects of the drought and recovery treatments: (1) depth distribution, (2)
166
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
reproductive mode, and (3) habitat specialisation. Data on lifetime fecundity of soil microarthropods such as Oribatida are not easy to obtain because it is difficult to measure, so literature data are scarce (see e.g., Norton, 1994). Furthermore, the ecology of rare species has seldom been well studied. Nevertheless, ecological characteristics of the more common European soil microarthropods have been documented often enough for the literature to provide useful information. We placed the microarthropods into ecological categories by using data from the literature and our own observations (N. Lindberg, pers obs.) (Table 2). Data on depth distribution were based on mean values from the control plots at Skogaby. Species were grouped as ‘‘surface living’’ (litter dwellers), ‘‘intermediate’’ (occurring evenly in the litter and humus layer) or ‘‘deep living’’ (living in the humus layer). Data on reproductive mode were mainly collected from Luxton (1981), Fjellberg (1980, 1998), Siepel (1995, 1996) and Maraun and Scheu (2000). If both sexual and parthenogenetic reproduction occurs in the same species, as in Mesaphorura macrochaeta (Fjellberg, 1998), the most common mode was assumed. For some species, data from related species of the same genus were used (Table 2). As a measure of habitat specialisation, information on each species’ occurrence in seven different habitats was collected from the literature. The habitats were (1) coniferous forest, (2) deciduous forest, (3) shrub heathland, (4) grassland, (5) ruderal areas, (6) cultivated fields and (7) tree trunks, stones and walls. A high value implies a large habitat niche width, but occurrence in a certain habitat was not registered if the literature indicated only occasional presence. The main sources for the microarthropod habitat data were ecological publications (Table 2). 2.5. Data treatment and statistical analysis The treatments drought, recovery and control were randomised within blocks and the three or four sampling occasions were regarded as repeated measurements. Rare species, missing from at least two of the years 1996–1999, or recorded as single specimens during two or more years, were excluded from further analyses, as were species where the determination was suspected to have been uncertain. Among the commonest species, the collembolans
Anurophorus septentrionalis and A. laricis were excluded due to possible confusion in the samples from 1999. The species-wise abundances were statistically analysed with a repeated measures ANOVA using autoregressive covariance structure of the order one, AR(1). All analyses of abundances were done with ln(x + 0.5) transformed data. Significant effects (P < 0.05) of treatment and year were further investigated using pairwise t-tests. All species that showed significant negative effects of drought compared to the control plots at the start of the study (1996) were considered ‘‘negatively affected’’ and chosen for subsequent analyses of recovery. Species that showed non-significant or positive effects of drought were considered ‘‘tolerant’’. Population recovery 1 year (1997) or 3 years (1999) after the cessation of the drought treatment was evaluated in the recovery plots as ‘‘early’’ and ‘‘late’’ recovery, respectively. A species was considered to have recovered if mean abundances in the recovery plots were at least 60% of the control abundances, and its abundances no longer differed significantly between the treatments. However, the population development in the recovery treatment compared to the drought treatment was also taken into account as non-significant differences were sometimes due to decreases in the control plots rather than to increases in the recovery plots. The samples from 1998 were not used because of the late start of the drought treatment combined with the unusually high precipitation in early summer that occurred this year (SMHI, 1998; Lindberg and Bengtsson, unpubl.). These samples were, however, retained in the ANOVA because samples from the same plots, but successive years, are dependent. The species were then placed in categories, first according to their drought and then to their recovery responses. Differences in the ecological characteristics between the categories were tested with Student’s t-tests (continuous variables) or G-tests (discrete variables). SAS for Windows, version 8.2 (SAS Institute, 2000) was used for most statistical tests. We decided not to apply any Bonferroni corrections of the significance levels, despite the fact that we performed many tests of species traits versus recovery. This is because we regard our study as exploratory, rather than testing a priori hypotheses. Only further studies of other ecosystems can show the generality of our results.
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
167
Table 2 Ecological characteristics of the oribatid and collembolan species at Skogaby included in the species-wise analyses 1996–1999 (see text) Species
Depth distribution at Skogaby
Habitat width (1–7)
Reproductive mode
Negatively affected: Oribatida Liochthonius sp. Phthiracarus cf. borealis Atropacarus striculus Nothrus silvestris Camisia biurus Camisia spinifer Nanhermannia coronata Cepheus cepheiformis Adoristes ovatus Tectocepheus velatus Suctobelbella spp. Oppiella nova Chamobates borealis Minunthozetes semirufus Eupelops torulosus Parachipteria punctata
I Sf I I Sf Sf I Sf Sf Sf D I Sf Sf Sf Sf
– 3 4 5 4 5 3 3 3 7 5 7 3 6 3 4
P S S P P P P – S P P P S S S S
22
Collembola Friesea mirabilis Neanura muscorum Willemia anophthalma Protaphorura pseudovanderdrifti Micraphorura absoloni Mesaphorura macrochaeta Isotomiella minor Isotoma viridis Isotoma notabilis
I Sf I I I D I Sf I
5 4 6 4 3 6 5 6 7
1,8,14,23,26,31
S P P S P P P S P
27
Sf Sf
3 4
2,3,4,13,20,28,34,35
S S
20
Oribatida Palaeacarus hystricinus Brachychthoniidae sp. Rhysotritia duplicata Porobelba spinosa+ Carabodes femoralis Carabodes labyrinthicus Zygoribatula exilis+ Oribatula tibialis Scheloribates pallidulus Hemileius initialis Ceratozetella thienemanni Oribatella calcarata
D D D I Sf Sf Sf Sf Sf Sf D Sf
2 – 4 4 3 6 2 5 4 5 1 2
35
Collembola Xenylla brevicauda
Sf
1
6,8
2,3,4,13,20,28,34,35 2,23,25,35 3,13,16,20,25,28,34,35 35 35 25,35 13,35 13,20,34,35 10,13,16,19,20,28,29,34,35 28,35 2,10,16,19,20,23,28,33,34,35,36 34,35 27,29,35,36 35 35
1,8,23,26,31 1,7,8,17,23,24 8 1,6,8 7,8,24,26 1,14,23,24,25 1,6,9,14,17,18,21,23,26,31,33,37 1,6,7,17,23,24,30,31,33,37
20 27 20 5 20 22
20 20 22
35
20 27 20 20
15
35 13,35 11,13,29,35 11,29,30,35 13,23,30,34,35 35 13,20,23,35,36 35 13,35
38 27 27
(C)
27 27 27
20
P 22 P 22 P 20 S 20 S*19 S 19 – S 20 S*20 S 19 S*20 S 20
–
(B)
27
Tolerant:
4,16,35
(A)
20
Weak negative effects (P < 0.10): Oribatida Phthiracarus cf. laevigatus Eupelops acromios
Remarks
(D)
168
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
Table 2 (Continued ) Species Paratullbergia callipygos Folsomia quadrioculata Lepidocyrtus cyaneus Lepidocyrtus lignorum Orchesella flavescens Orchesella bifasciata Entomobrya nivalis Entomobrya corticalis+
Depth distribution at Skogaby I Sf Sf Sf Sf Sf Sf Sf
Habitat width (1–7) 3 6 7 5 3 3 5 3
1,8,26 1,6,7,14,21,23,25,26,31 1,6,9,17,18,21,23,26,31 1,6,7,23,24,25,31,32,37 6,26,31 1,6 1,6,18,23,26,30,31,32 1,6,31
Reproductive mode
Remarks
27
P S 27 S*27 S*27 S 12 S*12 S 38 S 38
Sf = surface living, I = intermediate, D = deep living, P = parthenogenetic, S = sexually reproducing. + = species showing a significant positive response to drought (P < 0.05), – = data lacking, * = data based on species from the same genus. References; see footnotes.; Remarks: (A) Most of them S. similis and S. acutidens, (B) Sexual reproduction rarely (Hopkin, 1997), (C) Sexual reproduction rarely (Fjellberg, 1998), (D) Sexual reproduction rarely (Hopkin, 1997).; References used; for full reference account see reference list.; 1. Axelsson et al., 1984; 2. Beckmann, 1988; 3. Behan-Pelletier and Hill, 1983; 4. Berg, 1991; 5. Colloff, 1993; 6. Fjellberg, 1980; 7. Fjellberg, 1994; 8. Fjellberg, 1998; 9. Frampton et al., 2001; 10. Franchini and Rockett, 1996; 11. Gjelstrup and Søchting, 1984; 12. Hale, 1965; 13. Hammer, 1972; 14. Hansson et al., 1990; 15. Hopkin, 1997; 16. Hu¨ lsmann and Wolters, 1998; 17. Hutson, 1980; 18. Joosse, 1969; 19. Koehler, 1998; 20. Luxton, 1981; 21. Mebes and Filser, 1997; 22. Norton et al., 1993; 23. Persson and Lohm, 1977; 24. Petersen, 1995; 25. Schaefer and Schauermann, 1990; 26. Shaw, 1997; 27. Siepel, 1995; 28. Skubala, 1995; 29. Smrz and Kocourkova, 1999; 30. Steiner, 1994; 31. Tamm, 1986; 32. Verhoef and Van Selm, 1983; 33. Wanner and Dunger, 2002; 34. Webb, 1994; 35. Weigmann and Kratz, 1981; 36. Whelan, 1978; 37. Winkler and Kampichler, 2000; 38. M. Berg, personal communications.
3. Results In total, 74 species of microarthropods (47 Oribatida and 27 Collembola) were found, but 26 species (17 of which were Oribatida) had to be excluded from further analyses for reasons given above. Of the remaining 48 species, 25 showed significant negative effects of drought in 1996 and were considered negatively affected, whereas 21 showed non-significant or positive effects and were considered tolerant (Table 2). Two species of Oribatida and one of Collembola showed a positive effect. Two oribatids, Phthiracarus cf. laevigatus and Eupelops acromios, showed near-significant negative effects of drought (P < 0.06) but were excluded to ensure that our tests were conservative. 3.1. Differences in drought sensitivity No differences in depth preference of the negatively affected species compared with the tolerant species were found for Oribatida (G-test; P > 0.10), but among the Collembola, surface-living species were more common among the tolerant species (P < 0.05) than were species living deeper in the soil. There were proportionally more parthenogenetic species among the negatively affected Collembola compared with the tolerant species (Fig. 1a; G-test, P
< 0.05). This trend was also seen in Oribatida, although it was not statistically significant (Fig. 1b; P > 0.10). Differences in habitat width between negatively affected species and tolerant species were not statistically significant for either Oribatida or Collembola (ttests; P > 0.10) but, on average, tolerant species had more narrow habitat preferences in both groups (Fig. 1c). The difference was close to significance if both groups were pooled before the test (P = 0.07). 3.2. Differences in recovery rates Of the 25 collembolan and oribatid species that showed negative effects of drought at the start of the study, 14 species recovered during the studied period between 1997 and 1999 (Table 3). The populations of eight species did not recover during the studied period and three species were impossible to evaluate because their population fluctuations were too complex (Table 3). Among the negatively affected species that recovered, the Collembola showed significantly faster population recovery than the Oribatida (G-test; P < 0.05). Already in 1997, the densities of eight species of Collembola had reached control levels in the recovery plots, compared to none of the Oribatida (Table 3). Two years later, six species of Collembola and six of
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
169
Oribatida had recovered (Table 3). This year (1999) the populations of two collembolan species that seemed to have recovered in 1997 had dropped to levels significantly lower than in the control. The species that did not recover included seven Oribatida and one Collembola (Table 3). The proportion of Oribatida was significantly higher among the species that did not recover than among the 14 species that had recovered (G-test; P < 0.05). Only one collembolan species out of the nine that showed negative effects of drought, Friesea mirabilis, showed no signs of a population recovery (Table 3). Therefore, no further comparisons between recovering and non-recovering collembolan groups were made. For Oribatida, the depth preference did not differ between the species that recovered, and those that did not (P > 0.10), nor did the average habitat niche width (P > 0.10). However, there were relatively more parthenogenetic oribatid species among the species that recovered (Fig. 2), than among those which did not recover, and this difference was close to statistical significance (G-test; P < 0.10). Because of the limited number of species, comparisons between species with early and late population recovery could only be made after pooling the data for Collembola and Oribatida. No significant difference in depth preference or reproductive mode of the species was found between the groups (P > 0.10). Also, habitat niche width did not differ between the species that showed a fast population recovery and those that recovered later (P > 0.10), although species with larger habitat niche widths seemed to recover faster on average (Fig. 3).
4. Discussion Our results suggest that there were differences in ecological characteristics between negatively affected species and tolerant species, even though the communities in the small drought plots may have been buffered by short-range dispersal from the surround-
Fig. 1. Reproductive modes and habitat niche width for the negatively affected and drought-tolerant collembolan and oribatid species at Skogaby. (a) Reproductive modes of the collembolan species.
(b) Reproductive modes of the oribatid species. The differences in relative frequency of the reproductive modes were examined for each group by a G-test. (c) Mean habitat niche width (with S.E. bars) of the species of Oribatida (n = 26) and Collembola (n = 18). The differences in mean habitat niche width were examined for each group by a t-test.
170
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
Table 3 Population recovery of Oribatida and Collembola showing negative abundance effects of drought at Skogaby 1996–1999 Species
Difference CD (1996) P
Oribatida Liochthonius sp. Phthiracarus cf. borealis Atropacarus striculus Nothrus silvestris Camisia biurus Camisia spinifer Nanhermannia coronata Cepheus cepheiformis Adoristes ovatus Tectocepheus velatus Suctobelbella spp. Oppiella nova Chamobates borealis Minunthozetes semirufus Eupelops torulosus Parachipteria punctate
0.001 0.008 <0.001 0.026 0.010 0.040 <0.001 0.014 <0.001 <0.001 <0.001 0.025 0.005 0.025 0.003 <0.001
Collembola Friesea mirabilis Neanura muscorum Willemia anophthalma Protaphorura pseudovanderdrifti Micraphorura absoloni Mesaphorura macrochaeta Isotomiella minor Isotoma viridis Isotoma notabilis
0.011 <0.001 <0.001 0.014 <0.001 <0.001 <0.001 <0.001 0.009
‘‘Early recovery’’ after 1 year (1997)
‘‘Late recovery’’ after 3 years (1999)
Pop. recovery 1997–1999
?
? No No No Yes ? Yes Yes No No Yes Yes No ? Yes No
X ? X X
X X ? X
X X X X X X X X
X X X X X X
No Yes Yes Yes Yes Yes Yes Yes Yes
Significant treatment differences (P < 0.05) between drought and control plots are indicated. C = control, D = drought. X = population recovery, ? = evaluation not possible.
ings. Both the Oribatida and Collembola showed tendencies for the proportion of sexually reproducing species to be higher among the tolerant than among the negatively affected species. There also tended to be fewer deep-living species among the tolerant Collembola. Our data do not suggest that generalist species were less affected by the long-term drought disturbance than specialists. On the contrary, we found a trend that tolerant species of Oribatida and Collembola had narrower habitat preferences than negatively affected species. These differences between the groups of negatively affected and tolerant species can be explained by the presence of a number of drought-resistant collembolan and oribatid species with narrow habitat preferences. These species frequently occur in lichens and moss cushions on tree trunks, stones and buildings. They were mostly
rare in the control plots (Lindberg et al., 2002) but increased substantially during the drought treatment. In the Collembola, these species seemed to be phylogenetically correlated as they often belonged to the Entomobryidae: relatively large, surface-living, sexually reproducing species. Our results also suggest that a diverse species pool is important for maintaining abundances of functional groups in response to a disturbance, and the potential importance of rare species in maintaining ecosystem functions, both during and following disturbances (Walker, 1995; Elmqvist et al., 2003). We also found differences between the species whose populations recovered quickly, and those that were slower to recover. Collembolan species were quicker to recover than the Oribatida after the drought treatment had ceased. Also, among the species that did
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
Fig. 2. Reproductive modes of the negatively affected Oribatida species at Skogaby showing population recovery, or no recovery, within three years of the drought treatment ceasing. The difference in relative frequency of the reproductive modes was close to statistical significance according to a G-test (P < 0.10).
not recover at all, oribatids were over-represented. This finding is similar to the results of other post-disturbance studies of soil fauna (Strojan, 1978; Hutson, 1980; Lucarotti, 1981; Wanner and Dunger, 2002, but see also Luxton, 1982). Oribatids generally seem to have ‘‘K-selected’’ type traits, and more so than collembolans
Fig. 3. Mean habitat niche widths (with S.E. bars) of the negatively affected species of Collembola and Oribatida showing a fast population recovery (n = 8) and a slow recovery (n = 12), respectively (see text). The difference in mean habitat niche width was not significant, according to a t-test.
171
(Norton, 1994). However, other characteristics such as dispersal ability and habitat specialisation may also have affected the rate of recovery. We did not find strong support for the hypothesis that deep-living species would recover before surfaceliving ones (postulated on the grounds that they are better at avoiding drought, small and able to find sheltered microsites in the soil). However, the number of species available for this comparison was very small, and five of the six collembolans that recovered quickly were in fact from intermediate or deep soil layers. This pattern did not change if another measure that is indirectly related to depth preference, body size, was used instead. Mean body length did not differ between recovering and non-recovering species (results not shown). Even though negative effects of drought were more commonly found among parthenogenetic species, populations of such species also tended to recover more quickly. The reproductive mode may be crucial for successful colonisation. Parthenogenesis in microarthropods is assumed to be an advantageous trait for colonisers (Norton, 1994), as it may facilitate the establishment of populations from very few individuals. Indeed, the reproductive mode seemed to be important for recovery, as asexual reproduction was more common among the oribatid species that recovered than among those that did not. Among the Collembola, six of the eight species that showed quick population recovery were parthenogenetic. We used a coarse classification of habitat niche width due to a lack of detailed data on microhabitat preferences. Nevertheless, among the negatively affected species, those with narrow habitat preferences had a tendency to recover more slowly after drought. This finding is similar to results of studies by, for ˚ s et al. example, Baur and Bengtsson (1987) and A (1992) who showed that species with broader habitat niches tended to colonise islands earlier than habitat specialists. Soils are spatially heterogeneous habitats, and an arthropod’s chances of finding a vacant habitat niche and establishing itself upon arrival increase if the species has a wide niche preference. It is not clear whether recovery after drought depended mainly on the survival of remnant populations, or recolonisation from outside the plots. Comparisons with other faunal groups with different dispersal abilities indicate that dispersal ability
172
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
probably does affect the speed of recovery of soil faunal populations at the site since collembolans, and to some extent mesostigmatid mites, appeared to recover faster than oribatids (Lindberg and Bengtsson, unpublished). Furthermore, dispersal has been shown to be an important factor in maintenance of the diversity of microarthropods in isolated patches (Gonzalez and Chaneton, 2002). Conflicting evidence has been published about whether Oribatida or Collembola have higher dispersal ability in general (e.g., Hutson, 1980; Steiner, 1995, versus Ojala and Huhta, 2001). However, although measurements on dispersal rate are lacking for many soil faunal species, frequent occurrence in disturbed or newly created habitats should indicate good dispersal ability. Taxa known to be early colonisers, such as Mesaphorura sp. and Oppiella nova (Hutson, 1980; Koehler, 1998; Wanner and Dunger, 2002), were among the first Collembola and Oribatida, respectively, to show population recovery after the drought (Table 3). On the other hand, Tectocepheus velatus, another early oribatid coloniser (e.g., Maraun and Scheu, 2000), did not recover during the three-year study. Food resources may also influence succession if they enable e.g., faster development, higher oviposition rate and higher fecundity (Norton, 1994; Hopkin, 1997). However, we believe that changes in food resources are unlikely to have strongly affected the responses studied here, since no significant differences were found between the responses of fungal grazers and fungal browsers among the Oribatida (not shown), and gut-enzyme analyses indicate that the majority of collembolan species have similar feeding habits (M. Berg, pers. comm.). A number of factors may have influenced our results, complicating their interpretation. For instance, disturbance-sensitive species and species with poor dispersal ability may have been rare, or absent, at the site because the stand history selected against them. The relatively small number of microarthropod species that could be used for the statistical analyses, in spite of the large number of species at the site, imposed further limitations. In addition, because of phylogenetic correlations, certain characteristics were often combined in the same species, making their effects difficult to separate. Nevertheless, our results indicate that habitat specialisation and dispersal ability, and not only traits along the often criticized
‘‘r-K continuum’’ (Stearns, 1992), can strongly influence patterns of post-disturbance succession in soil fauna.
5. Conclusions We found that life-history traits and ecological characteristics seem to influence the responses of oribatid and collembolan species to drought disturbance. Parthenogenetic species and habitat generalist species were, on average, more negatively affected by the drought treatment than sexually reproducing species and species with narrow habitat preferences, respectively, but parthenogenetic species also seemed to recover faster after the disturbance. However, we could not conclusively show that the differences between oribatid and collembolan responses depended on the specific traits or characteristics studied. We believe that differences in dispersal rates between the two groups are likely to be strongly correlated with the differences in their population responses. However, estimates of dispersal rates of soil faunal species are very rare in the literature. Improved knowledge of such rates is needed to increase our understanding of soil faunal responses to disturbances, and may enable us to predict the impact of new disturbance regimes. Lasting effects of the drought were more common in Oribatida than in Collembola. Thus, the threat to Oribatida posed by extreme climate-related events and changes in land use may be greater than for other soil arthropods, as shown by the relatively slow population recovery of many oribatid species.
Acknowledgements We are grateful to Tryggve Persson for valuable support and sampling assistance, to Kerstin Ahlstro¨ m and Anna Carlsson for lab assistance, and to Birgitta Vegerfors-Persson for statistical advice. John Blackwell is thanked for a linguistic revision. Ulf Johansson was responsible for the maintenance of the experiment. The studies were supported by grants from the European Commission (ENV4-CT95-0027) as part of the GLOBIS project. The maintenance of the experiment was financed by the Faculty of Forestry, Swedish University of Agricultural Sciences.
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
References Alavi, G., 1999. Climate, Leaf Area, Soil Moisture and Tree Growth in Spruce Stands in SW Sweden – Field Experiments and Modelling. PhD thesis, Agraria 175, Swedish University of Agricultural Sciences, Uppsala. ˚ s, S., Bengtsson, J., Ebenhard, T., 1992. Archipelagoes and A theories of insularity. In: Hansson, L. (Ed.), Ecological Principles of Nature Conservation. Applications in Temperate and Boreal Environments, Elsevier Applied Science, London, pp. 201–251. Axelsson, B., Lohm, U., Persson, T., 1984. Enchytraeids, lumbricids and soil arthropods in a northern decidious woodland – a quantitative study. Holarctic Ecology 7, 91–103. Baker, H.G., 1955. Self-compatibility and establishment after longdistance dispersal. Evolution 9, 347–349. Baur, B., Bengtsson, J., 1987. Colonizing ability in land snails on Baltic uplift archipelagos. J. Biogeogr. 14, 329–341. Beckmann, M., 1988. Die Entwicklung der Bodenfauna ¨ kosystems und ihre Beeinflussung durch eines Ruderal-O Rekultivierung: 1. Oribatiden (Acari: Oribatei). Pedobiologia 31, 391–408. Behan-Pelletier, V.M., Hill, S.B., 1983. Feeding habits of sixteen species of Oribatei (Acari) from an acid peat bog, Glenamoy, Ireland. Rev. Ecol. Biol. Sol. 20, 221–267. Bengtsson, G., Rundgren, S., 1988. The Gusum case: a brass mill and the distribution of soil Collembola. Can. J. Zool. 66, 1518– 1526. Berg, J., 1991. Distribution pattern and phenology of Phthiracaridae and Euphthiracaridae (Acari: Oribatei) in a beech forest. In: Dusbabek, F., Bukva, V. (Eds.), Modern Acarology, Academia, Prague and SPB Academic Publishing BV, The Hague, pp. 521– 527. Bergholm, J., Jansson, P.-E., Johansson, U., Majdi, H., Nilsson, L.O., Persson, H., Rosengren-Brinck, U., Wiklund, K., 1995. Air pollution, tree vitality and forest production – the Skogaby project. General description of a field experiment with Norway spruce in south Sweden. In: Nilsson, L.-O., Hu¨ ttl, R.F., Johansson, U.T., Mathy, P. (Eds.), Nutrient Uptake and Cycling in Forest Ecosystems. Ecosystems Research Report 21, European Commission, Luxembourg, pp. 69–87. Boyce, M.S., 1984. Restitution of r- and K-selection as a model of density-dependent natural selection. Ann. Rev. Ecol. Syst. 15, 427–447. Colloff, M.J., 1993. A taxonomic revision of the oribatid mite genus Camisia (Acari: Oribatida). J. Nat. Hist. 27, 1325–1408. Davis, B.N.K., 1986. Colonization of newly created habitats by plants and animals. J. Environ. Manage. 22, 361–371. Diamond, J.M., 1974. Colonization of exploded volcanic islands by birds: The supertramp strategy. Science 184, 803–806. Ebenhard, T., 1991. Colonization in metapopulations: a review of theory and observations. Biol. J. Linn. Soc. 42, 105–121. Elmqvist, T., Folke, C., Nystro¨ m, M., Peterson, G., Bengtsson, J., Walker, B., Norberg, J., 2003. Response diversity, ecosystem change, and resilience. Front. Ecol. Environ. 1, 488–494. Fjellberg, A., 1980. Identification keys to Norwegian Collembola. ˚ s, 152 pp. The Norwegian Entomological Society, A
173
Fjellberg, A., 1994. The Collembola of the Norwegian arctic islands. Norsk Polarinstitutt, medd. nr. 133, Oslo, 57 pp. Fjellberg, A., 1998. The Collembola of Fennoscandia and Denmark. I: Poduromorpha and Onychiuridae. Fauna Entomologica Scandinavica 35. EJ Brill, The Netherlands, 184 pp. Frampton, G.K., Van den Brink, P.J., Wratten, S.D., 2001. Diel activity patterns in an arable collembolan community. Appl. Soil Ecol. 17, 63–80. Franchini, P., Rockett, C.L., 1996. Oribatid mites as ‘‘indicator’’ species for estimating the environmental impact of conventional and conservation tilling practices. Pedobiologia 40, 217–225. Gjelstrup, P., Søchting, U., 1984. Oribatid mites (Acarina) dominant in some lichen and moss species of maritime rocks on Bornholm in the Baltic. In: Griffiths, D.A., Bowman, C.E. (Eds.), Acarology VI, vol. 1, John Wiley & Sons Ltd., Chichester, pp. 528– 533. Gonzalez, A., Chaneton, E.J., 2002. Heterotroph species extinction, abundance and biomass dynamics in an experimentally fragmented microecosystem. J. Anim. Ecol. 71, 594–602. Grubb, P.J., 1987. Some generalizing ideas about colonization and succession in green plants and fungi. In: Gray, A.J., Crawley, M.J., Edwards, P.J. (Eds.), Colonization, Succession and Stability. Blackwell Scientific Publications, Oxford, pp. 81–102. Hale, W.G., 1965. Observations on the breeding biology of Collembola. Pedobiologia 5, 146–152. Hammer, M., 1972. Microhabitats of oribatid mites on a Danish woodland floor. Pedobiologia 12, 412–423. Hansson, A.-C., Andre´ n, O., Bostro¨ m, S., Bostro¨ m, U., Clarholm, M., Lagerlo¨ f, J., Lindberg, T., Paustian, K., Pettersson, R., Sohlenius, B., 1990. Structure of the agroecosystem. In: Andre´ n, O., Lindberg, T., Paustian, K., Rosswall, T. (Eds.), Ecology of Arable Land – Organisms, Carbon and Nitrogen Cycling. Ecol. Bull. 40, Copenhagen, pp. 41–83. Hopkin, S., 1997. Biology of the Springtails (Insecta: Collembola). Oxford University Press, Oxford, 330 pp. Hoy, J.B., 1990. Long-term effects of lindane on the oribatid community in a pine plantation. Environ. Entomol. 19, 898–900. Hu¨ lsmann, A., Wolters, V., 1998. The effects of different tillage practices on soil mites, with particular reference to Oribatida. Appl. Soil Ecol. 9, 327–332. Hutson, B.R., 1980. Colonization of industrial reclamation sites by acari, collembola and other invertebrates. J. Appl. Ecol. 17, 255– 275. Joosse, E.N.G., 1969. Population structure of some surface dwelling collembola in a coniferous forest soil. Neth. J. Zool. 19, 621– 634. Koehler, H., 1998. Secondary succession of soil mesofauna: A thirteen year study. Appl. Soil Ecol. 9, 81–86. Lindberg, N., 2003. Soil Fauna and Global Change—Responses to Experimental Drought, Irrigation, Fertilisation and Soil Warming. PhD thesis, Silvestria 270, Swedish University of Agricultural Sciences, Uppsala. Lindberg, N., Bengtsson, J., Persson, T., 2002. Effects of experimental irrigation and drought on the composition and diversity of soil fauna in a coniferous stand. J. Appl. Ecol. 39, 924–936.
174
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
Loreau, M., Mouquet, N., Gonzalez, A., 2003. Biodiversity as spatial insurance in heterogeneous landscapes. Proc. Natl. Acad. Sci. U.S.A. 100, 12765–12770. Lucarotti, C.J., 1981. The effect of fire and forest regeneration on mesofauna populations and microfungal species in lichen woodland soils. McGill Subarctic Res. Paper 32, 7–26. Luxton, M., 1981. Studies on the oribatid mites of a Danish beech wood soil. IV. Developmental biology. Pedobiologia 21, 312– 340. Luxton, M., 1982. The ecology of some soil mites from coal shale tips. J. Appl. Ecol. 19, 427–442. MacArthur, R.H., Wilson, E.O., 1967. The Theory of Island Biogeography. Princeton University Press, New Jersey. Maraun, M., Scheu, S., 2000. The structure of oribatid mite communities (Acari, Oribatida): patterns, mechanisms and implications for future research. Ecography 23, 374–383. Mebes, K.-H., Filser, J., 1997. A method for estimating the significance of surface dispersal for population fluctuations of Collembola in arable land. Pedobiologia 41, 115–122. Niemi, R., Karppinen, E., Uusitalo, M., 1997. Catalogue of the Oribatida (Acari) of Finland. Acta Zool. Fenn. 207, 1–39. Nilsson, L.-O., 1997. Manipulation of conventional forest management practices to increase forest growth – results from the Skogaby project. For. Ecol. Manage. 91, 53–60. Norton, R.A., 1994. Evolutionary aspects of oribatid mite life histories and consequences for the origin of the Astigmata. In: Houck, M. (Ed.), Mites. Ecological and Evolutionary Analyses of Life-History Patterns. Chapman & Hall, New York, pp. 99–135. Norton, R.A., Kethley, J.B., Johnston, D.E., O’Connor, B.M., 1993. Phylogenetic perspectives on genetic systems and reproductive modes of mites. In: Wrensch, D.A., Ebbert, M.A. (Eds.), Evolution and Diversity of Sex Ratio in Insects and Mites. Chapman & Hall, New York, pp. 8–99. Ojala, R., Huhta, V., 2001. Dispersal of microarthropods in forest soil. Pedobiologia 45, 443–450. Persson, T., Lohm, U., 1977. Energetical significance of the annelids and arthropods in a Swedish grassland soil. Ecol. Bull., 23, Swedish Natural Science Research Council, Stockholm, pp. 56– 141. Petersen, H., 1995. Temporal and spatial dynamics of soil Collembola during secondary succession in Danish heathland. Acta Zool. Fenn. 196, 190–194. Ronce, O., Perret, F., Olivieri, I., 2000. Landscape dynamics and evolution of colonizer syndromes: interactions between reproductive effort and dispersal in a metapopulation. Evol. Ecol. 14, 233–260. SAS Institute, 2000. SAS/Stat User’s guide, Version 8. SAS Institute Inc., Cary, North Carolina. Schaefer, M., Schauermann, J., 1990. The soil fauna of beech forests: comparison between a mull and a moder soil. Pedobiologia 34, 299–314. Shaw, P.J.A., 1997. Post-fire successions of Collembola in lowland heaths in South-Eastern UK. Pedobiologia 41, 80–87.
Siepel, H., 1995. Applications of microarthropod life-history tactics in nature management and ecotoxicology. Biol. Fert. Soils. 19, 75–83. Siepel, H., 1996. Biodiversity of soil microarthropods: the filtering of species. Biodivers. Conserv. 5, 251–260. Simberloff, D., 1981. What makes a good island colonist? In: Denno, R.F., Dingle, H. (Eds.), Insect Life History Patterns. Springer, New York, pp. 195–205. Skubala, P., 1995. Moss mites (Acarina: Oribatida) on industrial dumps of different ages. Pedobiologia 39, 170–184. SMHI, 1998. Va¨ dera˚ ret 1998. Va¨ der och Vatten. SMHI, Norrko¨ ping, 8 pp. (in Swedish). Smrz, J., Kocourkova, J., 1999. Mite communities of two epiphytic lichen species (Hypogymnia physodes and Parmelia sulcata) in the Czech Republic. Pedobiologia 43, 385–390. Stearns, S.C., 1992. The Evolution of Life Histories, Oxford University Press, Oxford. Steiner, W.A., 1994. The influence of air pollution on moss-dwelling animals: 1. Methodology and composition of flora and fauna. Rev. Suisse Zool. 101, 533–556. Steiner, W.A., 1995. Influence of air pollution on moss-dwelling animals: 3. Terrestrial fauna, with emphasis on Oribatida and Collembola. Acarologia 36, 149–173. St. John, M.G., Bagatto, G., Behan-Pelletier, V., Lindquist, E.E., Shorthouse, J.D., Smith, I.M., 2002. Mite (Acari) colonization of vegetated mine tailings near Sudbury, Ontario. Canada Plant Soil. 245, 295–305. Strojan, C.L., 1978. The impact of zinc smelter emissions on forest litter arthropods. Oikos 31, 41–46. Tamm, J.C., 1984. Surviving long submergence in the egg stage – a successful strategy of terrestrial arthropods living on flood plains (Collembola, Acari, Diptera). Oecologia 61, 417–419. Tamm, J.C., 1986. Fu¨ nfja¨ hrige Collembolensukzession auf einem verbrannten Kiefern-waldboden in Niedersachsen (BRD). Pedobiologia 29, 113–127. Verhoef, H.A., Van Selm, A.J., 1983. Distribution and population dynamics of Collembola in relation to soil moisture. Holarctic Ecol. 6, 387–394. Walker, B., 1995. Conserving biological diversity through ecosystem resilience. Conserv. Biol. 9, 747–752. Wanner, M., Dunger, W., 2002. Primary immigration and succession of soil organisms on reclaimed opencast coal mining areas in Eastern Germany. Eur. J. Soil Biol. 38, 137–143. Webb, N.R., 1994. Post-fire succession of cryptostigmatid mites (Acari, Cryptostigmata) in a Calluna-heathland soil. Pedobiologia 38, 138–145. Weigmann, V.G., Kratz, W., 1981. Die deutschen Hornmilbenarten und ihre o¨ kologische Charakteristik. Zoologische Beitra¨ ge. 27, 459–489. Whelan, J., 1978. Acarine succession in grassland on cutaway raised bog. Sci. Proc. R. Dublin Soc. 6, 175–183. Winkler, H., Kampichler, C., 2000. Local and regional species richness in communities of surface-dwelling grassland Collembola: indication of species saturation. Ecography 23, 385–392.
Population responses of oribatid mites and collembolans after drought N. Lindberg*, J. Bengtsson Department of Ecology and Crop Production Science, Swedish University of Agricultural Sciences, P.O. Box 7043, SE-750 07 Uppsala, Sweden Received 7 April 2004; received in revised form 4 July 2004; accepted 9 July 2004
Abstract To compare the effects of a drought disturbance on species of Oribatida and Collembola, and subsequent recovery of their populations after the drought, we examined a Norway spruce, Picea abies, stand in south-western Sweden, where 6 years of experimentally induced summer droughts had resulted in major changes in the soil faunal communities. We followed the population densities during a 4-year period and sought correlations between the species’ drought responses and their ecological characteristics. Data on depth preference, habitat choice and reproductive mode were collected from the literature. Surfaceliving species, which tended to have narrow habitat width, were less negatively affected by the drought. However, among species showing negative population responses to drought, species with large habitat widths tended to recover faster after the drought. Furthermore, parthenogenesis was more common among the oribatid species that showed a population recovery than among those that did not. Overall, collembolan species recovered faster than oribatids, and among the species that did not recover, Oribatida were over-represented. No general differences in characteristics between oribatids and collembolans were observed that could explain their different responses. Possibly, traits other than those examined were more important, such as differences in dispersal rates between the two groups. # 2004 Elsevier B.V. All rights reserved. Keywords: Collembola; Colonisation; Disturbance; Life history; Oribatida; Population recovery
1. Introduction Many ecosystems are subject to disturbances of varying frequencies, such as fire, drought, flooding and, when under human management, events like * Corresponding author. Tel.: +46 18 67 24 20; fax: +46 18 67 28 90. E-mail address: [email protected] (N. Lindberg).
clear-cutting or tilling. To persist in such ecosystems, organisms either have to survive the disturbance, or recolonise disturbed patches from the surroundings. Hence, the traits that allow species to survive in dynamic landscapes are likely to be different from those allowing species to persist locally (Loreau et al., 2003) and may be related to the colonising ability of the species. In colonisation theory, life-history traits have often been used to explain colonisation patterns
0929-1393/$ – see front matter # 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.apsoil.2004.07.003
164
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
in plants and animals (Ebenhard, 1991). Good colonisers, sometimes equated with ‘‘r-selected species’’, are often characterised by early age at first reproduction, short life span, large clutches and parthenogenesis, whereas the opposite has been suggested to be common among late colonisers, ‘‘K-selected species’’ (see, for instance, Baker, 1955; MacArthur and Wilson, 1967; Boyce, 1984; Grubb, 1987). However, the pattern is far from clear-cut, as landscape dynamics or traits such as dispersal ability and niche specialisation may also influence colonisation processes (Diamond, 1974; Simberloff, 1981; ˚ s et al., Baur and Bengtsson, 1987; Ebenhard, 1991; A 1992; Ronce et al., 2000). Post-disturbance colonisation and succession of soil fauna have received considerable interest over the years, and studies have focused, among others, on man-made habitats and the impact of pesticides, pollution and fire (e.g., Davis, 1986; Bengtsson and Rundgren, 1988; Hoy, 1990; Webb, 1994; Skubala, 1995; Wanner and Dunger, 2002). In soil microarthropods, dispersal behaviour, lifespan and reproductive strategies are probably key factors regulating survival and colonisation after disturbances (Norton, 1994; Petersen, 1995; Siepel, 1995; Shaw, 1997; St. John et al., 2002). It has also been suggested that the ability to survive in sheltered microsites is important (Tamm, 1984; Hopkin, 1997; Shaw, 1997). However, there have been few serious attempts to link the impact of disturbance and subsequent succession to the lifehistory traits and other characteristics of the species involved, probably because of a lack of data on many species (but see Siepel, 1996; Maraun and Scheu, 2000). Thus, general conclusions have often been drawn without proper tests. Oribatid mites and collembolans share many similar features as they are mainly fungivorous, have similar body sizes, and are very numerous groups in the organic soil layers. However, the general lifehistory traits of Oribatida have been considered typical of K-selected species (Norton, 1994), whereas collembolan species exhibit wider variation in lifehistory traits (Norton, 1994; Hopkin, 1997). Therefore, the species-wise responses of Oribatida and Collembola after a disturbance probably differ. It has been argued that disturbances are the main structuring force for oribatid mite communities (Acari: Oribatida) (Maraun and Scheu, 2000). Also,
effects of disturbances on soil faunal communities are often stronger and more long-lived in oribatids than in collembolans (Strojan, 1978; Hutson, 1980; Lucarotti, 1981; Lindberg et al., 2002; Lindberg and Bengtsson, unpublished). This may be a result of general differences in life-history, or differences in dispersal ability and habitat specialisation between the groups. To study the effects of a drought disturbance and the subsequent recovery of Oribatida and Collembola, we examined a forest site in south-western Sweden where experimentally induced long-term summer droughts had caused major changes in the soil faunal community (Lindberg et al., 2002). One advantage of the site was that the species pool consisted of many relatively well-known microarthropods. Our aim was to examine the extent to which a few life history and ecological characteristics could explain variations in the impact of drought on the soil fauna, and their recovery patterns after the disturbance. The main questions we addressed were: (1) Will the populations of collembolan species recover faster than oribatid species, as they are considered less K-selected (Norton, 1994)? (2) Will deep-living species recover faster than species living closer to the surface, as they are more likely to have survived the disturbance due to their ability to penetrate deeper into the soil during drought? (3) Will populations of parthenogenetic species recover faster than sexually reproducing ones, as suggested by r-selection theory? (4) Will generalist species recover faster than species with narrower habitat preferences, as suggested by ˚ s et al. (1992) for fragmented habitats in e.g., A general?
2. Material and methods 2.1. Site description The field study was carried out during 1996–1999 in a homogeneous Norway spruce (Picea abies) stand situated at Skogaby, south-western Sweden (568330 N, 138130 E), 16 km from the coast at an altitude of 95– 115 m above sea level. The climate is maritime with a
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
mean annual precipitation of about 1150 mm, and an annual mean temperature of 7.5 8C. The stand was a second-generation forest on former heathland. It was 32 years old when the present study started and the trees were planted as 2-year-old seedlings in 1966. For more details on soil conditions and stand history, see Lindberg et al. (2002). The stand was very dense, and due to the closed canopy less than 2% of the incident light penetrated to the ground. No field or shrub layer existed and the under-storey consisted of sparse mosses with scattered tussocks of grass. No changes in vegetation cover on the plots were detected during the studied period. The soil conditions and stand type at Skogaby are typical for large areas of managed forest in southern Sweden (Bergholm et al., 1995). 2.2. Experimental design We used plots that were part of a large field experiment established in 1988 to study effects of water and nutrient availability on tree growth (Nilsson, 1997). It had a randomised block design with four blocks, each containing one replicate of each treatment. The present study examined the effects of two of the seven treatments, namely control (no treatment) and summer drought (created by roofs covering most of the plot area). Plot size was 45 m 45 m for the control plots and, initially (1990–1996) 45 m 22.5 m for the drought plots. A recovery treatment was then initiated by permanently removing roofs from half of each drought plot’s surface, giving drought plots and recovery plots, both measuring 22.5 m 22.5 m. The drought treatment started in 1990 and the recovery treatment started in autumn 1996. The plastic roofs were transparent and placed 1–1.5 m above ground. They had openings for tree trunks and for maintenance purposes, but prevented 70% of the throughfall on the plots from reaching the ground during April to September (Alavi, 1999). During the winter the roofs were removed and all throughfall could reach the ground (Table 1), resulting in a summer drought treatment (see Lindberg et al., 2002 for more details). Litter falling on the roofs was spread on the ground when the roofs were removed in October. During the period of roof coverage approximately 40% of the yearly litterfall occurred (L-O. Nilsson, pers. comm.).
165
Table 1 Duration of the summer drought treatment and data on timing and sampled soil surface during the soil faunal samplings between 1996 and 1999 Year
Roofs on
Roofs off
Sampling
Sampled surface/plot
1996 1997 1998 1999
Mid-April Mid-April Mid-July Early June
Early October Early October Mid-December Early October
Oct. 29 Oct. 20 Nov. 12 Nov. 4
3 dm2 4 dm2 4 dm2 3 dm2
2.3. Sampling and species identification The plots were sampled once every year, in late autumn 1996–1999, 14–34 days after roof removal (Table 1). Arthropods were collected by taking three or four systematically distributed 100-cm2 10-cmdeep subsamples of the litter and humus layers from each plot, and the layers were separated. The sampling points were evenly distributed within the plots, placed 7.5 m from the edge on the lines between the centre of each side and the mid-point of the plot. When sampling in drought plots we avoided areas closer than 1 m to tree trunks and gaps in the roof coverage. In the laboratory the subsamples were stored at 4 8C for 12 h before extraction in Tullgren dry funnels for 4 days. Due to practical constraints all subsamples from each plot and layer were pooled before extraction in 1996– 1997, but this was not done in 1998–1999. The samples from 1998 were later excluded from the present study as the late start of the drought treatment in this year probably affected the drought reference and complicated recovery analysis (see Lindberg and Bengtsson, unpublished). All adult individuals of Collembola and Oribatida were determined to the species level (see Lindberg et al., 2002). Oribatid nomenclature follows Niemi et al. (1997), while collembolan nomenclature follows Fjellberg (1998) for Poduromorpha, and Fjellberg (1980) for other groups. All data can be found in Lindberg (2003). 2.4. Life-history and ecological traits To examine how species recovery was related to life-history and ecological characteristics, the following traits were related to the effects of the drought and recovery treatments: (1) depth distribution, (2)
166
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
reproductive mode, and (3) habitat specialisation. Data on lifetime fecundity of soil microarthropods such as Oribatida are not easy to obtain because it is difficult to measure, so literature data are scarce (see e.g., Norton, 1994). Furthermore, the ecology of rare species has seldom been well studied. Nevertheless, ecological characteristics of the more common European soil microarthropods have been documented often enough for the literature to provide useful information. We placed the microarthropods into ecological categories by using data from the literature and our own observations (N. Lindberg, pers obs.) (Table 2). Data on depth distribution were based on mean values from the control plots at Skogaby. Species were grouped as ‘‘surface living’’ (litter dwellers), ‘‘intermediate’’ (occurring evenly in the litter and humus layer) or ‘‘deep living’’ (living in the humus layer). Data on reproductive mode were mainly collected from Luxton (1981), Fjellberg (1980, 1998), Siepel (1995, 1996) and Maraun and Scheu (2000). If both sexual and parthenogenetic reproduction occurs in the same species, as in Mesaphorura macrochaeta (Fjellberg, 1998), the most common mode was assumed. For some species, data from related species of the same genus were used (Table 2). As a measure of habitat specialisation, information on each species’ occurrence in seven different habitats was collected from the literature. The habitats were (1) coniferous forest, (2) deciduous forest, (3) shrub heathland, (4) grassland, (5) ruderal areas, (6) cultivated fields and (7) tree trunks, stones and walls. A high value implies a large habitat niche width, but occurrence in a certain habitat was not registered if the literature indicated only occasional presence. The main sources for the microarthropod habitat data were ecological publications (Table 2). 2.5. Data treatment and statistical analysis The treatments drought, recovery and control were randomised within blocks and the three or four sampling occasions were regarded as repeated measurements. Rare species, missing from at least two of the years 1996–1999, or recorded as single specimens during two or more years, were excluded from further analyses, as were species where the determination was suspected to have been uncertain. Among the commonest species, the collembolans
Anurophorus septentrionalis and A. laricis were excluded due to possible confusion in the samples from 1999. The species-wise abundances were statistically analysed with a repeated measures ANOVA using autoregressive covariance structure of the order one, AR(1). All analyses of abundances were done with ln(x + 0.5) transformed data. Significant effects (P < 0.05) of treatment and year were further investigated using pairwise t-tests. All species that showed significant negative effects of drought compared to the control plots at the start of the study (1996) were considered ‘‘negatively affected’’ and chosen for subsequent analyses of recovery. Species that showed non-significant or positive effects of drought were considered ‘‘tolerant’’. Population recovery 1 year (1997) or 3 years (1999) after the cessation of the drought treatment was evaluated in the recovery plots as ‘‘early’’ and ‘‘late’’ recovery, respectively. A species was considered to have recovered if mean abundances in the recovery plots were at least 60% of the control abundances, and its abundances no longer differed significantly between the treatments. However, the population development in the recovery treatment compared to the drought treatment was also taken into account as non-significant differences were sometimes due to decreases in the control plots rather than to increases in the recovery plots. The samples from 1998 were not used because of the late start of the drought treatment combined with the unusually high precipitation in early summer that occurred this year (SMHI, 1998; Lindberg and Bengtsson, unpubl.). These samples were, however, retained in the ANOVA because samples from the same plots, but successive years, are dependent. The species were then placed in categories, first according to their drought and then to their recovery responses. Differences in the ecological characteristics between the categories were tested with Student’s t-tests (continuous variables) or G-tests (discrete variables). SAS for Windows, version 8.2 (SAS Institute, 2000) was used for most statistical tests. We decided not to apply any Bonferroni corrections of the significance levels, despite the fact that we performed many tests of species traits versus recovery. This is because we regard our study as exploratory, rather than testing a priori hypotheses. Only further studies of other ecosystems can show the generality of our results.
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
167
Table 2 Ecological characteristics of the oribatid and collembolan species at Skogaby included in the species-wise analyses 1996–1999 (see text) Species
Depth distribution at Skogaby
Habitat width (1–7)
Reproductive mode
Negatively affected: Oribatida Liochthonius sp. Phthiracarus cf. borealis Atropacarus striculus Nothrus silvestris Camisia biurus Camisia spinifer Nanhermannia coronata Cepheus cepheiformis Adoristes ovatus Tectocepheus velatus Suctobelbella spp. Oppiella nova Chamobates borealis Minunthozetes semirufus Eupelops torulosus Parachipteria punctata
I Sf I I Sf Sf I Sf Sf Sf D I Sf Sf Sf Sf
– 3 4 5 4 5 3 3 3 7 5 7 3 6 3 4
P S S P P P P – S P P P S S S S
22
Collembola Friesea mirabilis Neanura muscorum Willemia anophthalma Protaphorura pseudovanderdrifti Micraphorura absoloni Mesaphorura macrochaeta Isotomiella minor Isotoma viridis Isotoma notabilis
I Sf I I I D I Sf I
5 4 6 4 3 6 5 6 7
1,8,14,23,26,31
S P P S P P P S P
27
Sf Sf
3 4
2,3,4,13,20,28,34,35
S S
20
Oribatida Palaeacarus hystricinus Brachychthoniidae sp. Rhysotritia duplicata Porobelba spinosa+ Carabodes femoralis Carabodes labyrinthicus Zygoribatula exilis+ Oribatula tibialis Scheloribates pallidulus Hemileius initialis Ceratozetella thienemanni Oribatella calcarata
D D D I Sf Sf Sf Sf Sf Sf D Sf
2 – 4 4 3 6 2 5 4 5 1 2
35
Collembola Xenylla brevicauda
Sf
1
6,8
2,3,4,13,20,28,34,35 2,23,25,35 3,13,16,20,25,28,34,35 35 35 25,35 13,35 13,20,34,35 10,13,16,19,20,28,29,34,35 28,35 2,10,16,19,20,23,28,33,34,35,36 34,35 27,29,35,36 35 35
1,8,23,26,31 1,7,8,17,23,24 8 1,6,8 7,8,24,26 1,14,23,24,25 1,6,9,14,17,18,21,23,26,31,33,37 1,6,7,17,23,24,30,31,33,37
20 27 20 5 20 22
20 20 22
35
20 27 20 20
15
35 13,35 11,13,29,35 11,29,30,35 13,23,30,34,35 35 13,20,23,35,36 35 13,35
38 27 27
(C)
27 27 27
20
P 22 P 22 P 20 S 20 S*19 S 19 – S 20 S*20 S 19 S*20 S 20
–
(B)
27
Tolerant:
4,16,35
(A)
20
Weak negative effects (P < 0.10): Oribatida Phthiracarus cf. laevigatus Eupelops acromios
Remarks
(D)
168
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
Table 2 (Continued ) Species Paratullbergia callipygos Folsomia quadrioculata Lepidocyrtus cyaneus Lepidocyrtus lignorum Orchesella flavescens Orchesella bifasciata Entomobrya nivalis Entomobrya corticalis+
Depth distribution at Skogaby I Sf Sf Sf Sf Sf Sf Sf
Habitat width (1–7) 3 6 7 5 3 3 5 3
1,8,26 1,6,7,14,21,23,25,26,31 1,6,9,17,18,21,23,26,31 1,6,7,23,24,25,31,32,37 6,26,31 1,6 1,6,18,23,26,30,31,32 1,6,31
Reproductive mode
Remarks
27
P S 27 S*27 S*27 S 12 S*12 S 38 S 38
Sf = surface living, I = intermediate, D = deep living, P = parthenogenetic, S = sexually reproducing. + = species showing a significant positive response to drought (P < 0.05), – = data lacking, * = data based on species from the same genus. References; see footnotes.; Remarks: (A) Most of them S. similis and S. acutidens, (B) Sexual reproduction rarely (Hopkin, 1997), (C) Sexual reproduction rarely (Fjellberg, 1998), (D) Sexual reproduction rarely (Hopkin, 1997).; References used; for full reference account see reference list.; 1. Axelsson et al., 1984; 2. Beckmann, 1988; 3. Behan-Pelletier and Hill, 1983; 4. Berg, 1991; 5. Colloff, 1993; 6. Fjellberg, 1980; 7. Fjellberg, 1994; 8. Fjellberg, 1998; 9. Frampton et al., 2001; 10. Franchini and Rockett, 1996; 11. Gjelstrup and Søchting, 1984; 12. Hale, 1965; 13. Hammer, 1972; 14. Hansson et al., 1990; 15. Hopkin, 1997; 16. Hu¨ lsmann and Wolters, 1998; 17. Hutson, 1980; 18. Joosse, 1969; 19. Koehler, 1998; 20. Luxton, 1981; 21. Mebes and Filser, 1997; 22. Norton et al., 1993; 23. Persson and Lohm, 1977; 24. Petersen, 1995; 25. Schaefer and Schauermann, 1990; 26. Shaw, 1997; 27. Siepel, 1995; 28. Skubala, 1995; 29. Smrz and Kocourkova, 1999; 30. Steiner, 1994; 31. Tamm, 1986; 32. Verhoef and Van Selm, 1983; 33. Wanner and Dunger, 2002; 34. Webb, 1994; 35. Weigmann and Kratz, 1981; 36. Whelan, 1978; 37. Winkler and Kampichler, 2000; 38. M. Berg, personal communications.
3. Results In total, 74 species of microarthropods (47 Oribatida and 27 Collembola) were found, but 26 species (17 of which were Oribatida) had to be excluded from further analyses for reasons given above. Of the remaining 48 species, 25 showed significant negative effects of drought in 1996 and were considered negatively affected, whereas 21 showed non-significant or positive effects and were considered tolerant (Table 2). Two species of Oribatida and one of Collembola showed a positive effect. Two oribatids, Phthiracarus cf. laevigatus and Eupelops acromios, showed near-significant negative effects of drought (P < 0.06) but were excluded to ensure that our tests were conservative. 3.1. Differences in drought sensitivity No differences in depth preference of the negatively affected species compared with the tolerant species were found for Oribatida (G-test; P > 0.10), but among the Collembola, surface-living species were more common among the tolerant species (P < 0.05) than were species living deeper in the soil. There were proportionally more parthenogenetic species among the negatively affected Collembola compared with the tolerant species (Fig. 1a; G-test, P
< 0.05). This trend was also seen in Oribatida, although it was not statistically significant (Fig. 1b; P > 0.10). Differences in habitat width between negatively affected species and tolerant species were not statistically significant for either Oribatida or Collembola (ttests; P > 0.10) but, on average, tolerant species had more narrow habitat preferences in both groups (Fig. 1c). The difference was close to significance if both groups were pooled before the test (P = 0.07). 3.2. Differences in recovery rates Of the 25 collembolan and oribatid species that showed negative effects of drought at the start of the study, 14 species recovered during the studied period between 1997 and 1999 (Table 3). The populations of eight species did not recover during the studied period and three species were impossible to evaluate because their population fluctuations were too complex (Table 3). Among the negatively affected species that recovered, the Collembola showed significantly faster population recovery than the Oribatida (G-test; P < 0.05). Already in 1997, the densities of eight species of Collembola had reached control levels in the recovery plots, compared to none of the Oribatida (Table 3). Two years later, six species of Collembola and six of
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
169
Oribatida had recovered (Table 3). This year (1999) the populations of two collembolan species that seemed to have recovered in 1997 had dropped to levels significantly lower than in the control. The species that did not recover included seven Oribatida and one Collembola (Table 3). The proportion of Oribatida was significantly higher among the species that did not recover than among the 14 species that had recovered (G-test; P < 0.05). Only one collembolan species out of the nine that showed negative effects of drought, Friesea mirabilis, showed no signs of a population recovery (Table 3). Therefore, no further comparisons between recovering and non-recovering collembolan groups were made. For Oribatida, the depth preference did not differ between the species that recovered, and those that did not (P > 0.10), nor did the average habitat niche width (P > 0.10). However, there were relatively more parthenogenetic oribatid species among the species that recovered (Fig. 2), than among those which did not recover, and this difference was close to statistical significance (G-test; P < 0.10). Because of the limited number of species, comparisons between species with early and late population recovery could only be made after pooling the data for Collembola and Oribatida. No significant difference in depth preference or reproductive mode of the species was found between the groups (P > 0.10). Also, habitat niche width did not differ between the species that showed a fast population recovery and those that recovered later (P > 0.10), although species with larger habitat niche widths seemed to recover faster on average (Fig. 3).
4. Discussion Our results suggest that there were differences in ecological characteristics between negatively affected species and tolerant species, even though the communities in the small drought plots may have been buffered by short-range dispersal from the surround-
Fig. 1. Reproductive modes and habitat niche width for the negatively affected and drought-tolerant collembolan and oribatid species at Skogaby. (a) Reproductive modes of the collembolan species.
(b) Reproductive modes of the oribatid species. The differences in relative frequency of the reproductive modes were examined for each group by a G-test. (c) Mean habitat niche width (with S.E. bars) of the species of Oribatida (n = 26) and Collembola (n = 18). The differences in mean habitat niche width were examined for each group by a t-test.
170
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
Table 3 Population recovery of Oribatida and Collembola showing negative abundance effects of drought at Skogaby 1996–1999 Species
Difference CD (1996) P
Oribatida Liochthonius sp. Phthiracarus cf. borealis Atropacarus striculus Nothrus silvestris Camisia biurus Camisia spinifer Nanhermannia coronata Cepheus cepheiformis Adoristes ovatus Tectocepheus velatus Suctobelbella spp. Oppiella nova Chamobates borealis Minunthozetes semirufus Eupelops torulosus Parachipteria punctate
0.001 0.008 <0.001 0.026 0.010 0.040 <0.001 0.014 <0.001 <0.001 <0.001 0.025 0.005 0.025 0.003 <0.001
Collembola Friesea mirabilis Neanura muscorum Willemia anophthalma Protaphorura pseudovanderdrifti Micraphorura absoloni Mesaphorura macrochaeta Isotomiella minor Isotoma viridis Isotoma notabilis
0.011 <0.001 <0.001 0.014 <0.001 <0.001 <0.001 <0.001 0.009
‘‘Early recovery’’ after 1 year (1997)
‘‘Late recovery’’ after 3 years (1999)
Pop. recovery 1997–1999
?
? No No No Yes ? Yes Yes No No Yes Yes No ? Yes No
X ? X X
X X ? X
X X X X X X X X
X X X X X X
No Yes Yes Yes Yes Yes Yes Yes Yes
Significant treatment differences (P < 0.05) between drought and control plots are indicated. C = control, D = drought. X = population recovery, ? = evaluation not possible.
ings. Both the Oribatida and Collembola showed tendencies for the proportion of sexually reproducing species to be higher among the tolerant than among the negatively affected species. There also tended to be fewer deep-living species among the tolerant Collembola. Our data do not suggest that generalist species were less affected by the long-term drought disturbance than specialists. On the contrary, we found a trend that tolerant species of Oribatida and Collembola had narrower habitat preferences than negatively affected species. These differences between the groups of negatively affected and tolerant species can be explained by the presence of a number of drought-resistant collembolan and oribatid species with narrow habitat preferences. These species frequently occur in lichens and moss cushions on tree trunks, stones and buildings. They were mostly
rare in the control plots (Lindberg et al., 2002) but increased substantially during the drought treatment. In the Collembola, these species seemed to be phylogenetically correlated as they often belonged to the Entomobryidae: relatively large, surface-living, sexually reproducing species. Our results also suggest that a diverse species pool is important for maintaining abundances of functional groups in response to a disturbance, and the potential importance of rare species in maintaining ecosystem functions, both during and following disturbances (Walker, 1995; Elmqvist et al., 2003). We also found differences between the species whose populations recovered quickly, and those that were slower to recover. Collembolan species were quicker to recover than the Oribatida after the drought treatment had ceased. Also, among the species that did
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
Fig. 2. Reproductive modes of the negatively affected Oribatida species at Skogaby showing population recovery, or no recovery, within three years of the drought treatment ceasing. The difference in relative frequency of the reproductive modes was close to statistical significance according to a G-test (P < 0.10).
not recover at all, oribatids were over-represented. This finding is similar to the results of other post-disturbance studies of soil fauna (Strojan, 1978; Hutson, 1980; Lucarotti, 1981; Wanner and Dunger, 2002, but see also Luxton, 1982). Oribatids generally seem to have ‘‘K-selected’’ type traits, and more so than collembolans
Fig. 3. Mean habitat niche widths (with S.E. bars) of the negatively affected species of Collembola and Oribatida showing a fast population recovery (n = 8) and a slow recovery (n = 12), respectively (see text). The difference in mean habitat niche width was not significant, according to a t-test.
171
(Norton, 1994). However, other characteristics such as dispersal ability and habitat specialisation may also have affected the rate of recovery. We did not find strong support for the hypothesis that deep-living species would recover before surfaceliving ones (postulated on the grounds that they are better at avoiding drought, small and able to find sheltered microsites in the soil). However, the number of species available for this comparison was very small, and five of the six collembolans that recovered quickly were in fact from intermediate or deep soil layers. This pattern did not change if another measure that is indirectly related to depth preference, body size, was used instead. Mean body length did not differ between recovering and non-recovering species (results not shown). Even though negative effects of drought were more commonly found among parthenogenetic species, populations of such species also tended to recover more quickly. The reproductive mode may be crucial for successful colonisation. Parthenogenesis in microarthropods is assumed to be an advantageous trait for colonisers (Norton, 1994), as it may facilitate the establishment of populations from very few individuals. Indeed, the reproductive mode seemed to be important for recovery, as asexual reproduction was more common among the oribatid species that recovered than among those that did not. Among the Collembola, six of the eight species that showed quick population recovery were parthenogenetic. We used a coarse classification of habitat niche width due to a lack of detailed data on microhabitat preferences. Nevertheless, among the negatively affected species, those with narrow habitat preferences had a tendency to recover more slowly after drought. This finding is similar to results of studies by, for ˚ s et al. example, Baur and Bengtsson (1987) and A (1992) who showed that species with broader habitat niches tended to colonise islands earlier than habitat specialists. Soils are spatially heterogeneous habitats, and an arthropod’s chances of finding a vacant habitat niche and establishing itself upon arrival increase if the species has a wide niche preference. It is not clear whether recovery after drought depended mainly on the survival of remnant populations, or recolonisation from outside the plots. Comparisons with other faunal groups with different dispersal abilities indicate that dispersal ability
172
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
probably does affect the speed of recovery of soil faunal populations at the site since collembolans, and to some extent mesostigmatid mites, appeared to recover faster than oribatids (Lindberg and Bengtsson, unpublished). Furthermore, dispersal has been shown to be an important factor in maintenance of the diversity of microarthropods in isolated patches (Gonzalez and Chaneton, 2002). Conflicting evidence has been published about whether Oribatida or Collembola have higher dispersal ability in general (e.g., Hutson, 1980; Steiner, 1995, versus Ojala and Huhta, 2001). However, although measurements on dispersal rate are lacking for many soil faunal species, frequent occurrence in disturbed or newly created habitats should indicate good dispersal ability. Taxa known to be early colonisers, such as Mesaphorura sp. and Oppiella nova (Hutson, 1980; Koehler, 1998; Wanner and Dunger, 2002), were among the first Collembola and Oribatida, respectively, to show population recovery after the drought (Table 3). On the other hand, Tectocepheus velatus, another early oribatid coloniser (e.g., Maraun and Scheu, 2000), did not recover during the three-year study. Food resources may also influence succession if they enable e.g., faster development, higher oviposition rate and higher fecundity (Norton, 1994; Hopkin, 1997). However, we believe that changes in food resources are unlikely to have strongly affected the responses studied here, since no significant differences were found between the responses of fungal grazers and fungal browsers among the Oribatida (not shown), and gut-enzyme analyses indicate that the majority of collembolan species have similar feeding habits (M. Berg, pers. comm.). A number of factors may have influenced our results, complicating their interpretation. For instance, disturbance-sensitive species and species with poor dispersal ability may have been rare, or absent, at the site because the stand history selected against them. The relatively small number of microarthropod species that could be used for the statistical analyses, in spite of the large number of species at the site, imposed further limitations. In addition, because of phylogenetic correlations, certain characteristics were often combined in the same species, making their effects difficult to separate. Nevertheless, our results indicate that habitat specialisation and dispersal ability, and not only traits along the often criticized
‘‘r-K continuum’’ (Stearns, 1992), can strongly influence patterns of post-disturbance succession in soil fauna.
5. Conclusions We found that life-history traits and ecological characteristics seem to influence the responses of oribatid and collembolan species to drought disturbance. Parthenogenetic species and habitat generalist species were, on average, more negatively affected by the drought treatment than sexually reproducing species and species with narrow habitat preferences, respectively, but parthenogenetic species also seemed to recover faster after the disturbance. However, we could not conclusively show that the differences between oribatid and collembolan responses depended on the specific traits or characteristics studied. We believe that differences in dispersal rates between the two groups are likely to be strongly correlated with the differences in their population responses. However, estimates of dispersal rates of soil faunal species are very rare in the literature. Improved knowledge of such rates is needed to increase our understanding of soil faunal responses to disturbances, and may enable us to predict the impact of new disturbance regimes. Lasting effects of the drought were more common in Oribatida than in Collembola. Thus, the threat to Oribatida posed by extreme climate-related events and changes in land use may be greater than for other soil arthropods, as shown by the relatively slow population recovery of many oribatid species.
Acknowledgements We are grateful to Tryggve Persson for valuable support and sampling assistance, to Kerstin Ahlstro¨ m and Anna Carlsson for lab assistance, and to Birgitta Vegerfors-Persson for statistical advice. John Blackwell is thanked for a linguistic revision. Ulf Johansson was responsible for the maintenance of the experiment. The studies were supported by grants from the European Commission (ENV4-CT95-0027) as part of the GLOBIS project. The maintenance of the experiment was financed by the Faculty of Forestry, Swedish University of Agricultural Sciences.
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
References Alavi, G., 1999. Climate, Leaf Area, Soil Moisture and Tree Growth in Spruce Stands in SW Sweden – Field Experiments and Modelling. PhD thesis, Agraria 175, Swedish University of Agricultural Sciences, Uppsala. ˚ s, S., Bengtsson, J., Ebenhard, T., 1992. Archipelagoes and A theories of insularity. In: Hansson, L. (Ed.), Ecological Principles of Nature Conservation. Applications in Temperate and Boreal Environments, Elsevier Applied Science, London, pp. 201–251. Axelsson, B., Lohm, U., Persson, T., 1984. Enchytraeids, lumbricids and soil arthropods in a northern decidious woodland – a quantitative study. Holarctic Ecology 7, 91–103. Baker, H.G., 1955. Self-compatibility and establishment after longdistance dispersal. Evolution 9, 347–349. Baur, B., Bengtsson, J., 1987. Colonizing ability in land snails on Baltic uplift archipelagos. J. Biogeogr. 14, 329–341. Beckmann, M., 1988. Die Entwicklung der Bodenfauna ¨ kosystems und ihre Beeinflussung durch eines Ruderal-O Rekultivierung: 1. Oribatiden (Acari: Oribatei). Pedobiologia 31, 391–408. Behan-Pelletier, V.M., Hill, S.B., 1983. Feeding habits of sixteen species of Oribatei (Acari) from an acid peat bog, Glenamoy, Ireland. Rev. Ecol. Biol. Sol. 20, 221–267. Bengtsson, G., Rundgren, S., 1988. The Gusum case: a brass mill and the distribution of soil Collembola. Can. J. Zool. 66, 1518– 1526. Berg, J., 1991. Distribution pattern and phenology of Phthiracaridae and Euphthiracaridae (Acari: Oribatei) in a beech forest. In: Dusbabek, F., Bukva, V. (Eds.), Modern Acarology, Academia, Prague and SPB Academic Publishing BV, The Hague, pp. 521– 527. Bergholm, J., Jansson, P.-E., Johansson, U., Majdi, H., Nilsson, L.O., Persson, H., Rosengren-Brinck, U., Wiklund, K., 1995. Air pollution, tree vitality and forest production – the Skogaby project. General description of a field experiment with Norway spruce in south Sweden. In: Nilsson, L.-O., Hu¨ ttl, R.F., Johansson, U.T., Mathy, P. (Eds.), Nutrient Uptake and Cycling in Forest Ecosystems. Ecosystems Research Report 21, European Commission, Luxembourg, pp. 69–87. Boyce, M.S., 1984. Restitution of r- and K-selection as a model of density-dependent natural selection. Ann. Rev. Ecol. Syst. 15, 427–447. Colloff, M.J., 1993. A taxonomic revision of the oribatid mite genus Camisia (Acari: Oribatida). J. Nat. Hist. 27, 1325–1408. Davis, B.N.K., 1986. Colonization of newly created habitats by plants and animals. J. Environ. Manage. 22, 361–371. Diamond, J.M., 1974. Colonization of exploded volcanic islands by birds: The supertramp strategy. Science 184, 803–806. Ebenhard, T., 1991. Colonization in metapopulations: a review of theory and observations. Biol. J. Linn. Soc. 42, 105–121. Elmqvist, T., Folke, C., Nystro¨ m, M., Peterson, G., Bengtsson, J., Walker, B., Norberg, J., 2003. Response diversity, ecosystem change, and resilience. Front. Ecol. Environ. 1, 488–494. Fjellberg, A., 1980. Identification keys to Norwegian Collembola. ˚ s, 152 pp. The Norwegian Entomological Society, A
173
Fjellberg, A., 1994. The Collembola of the Norwegian arctic islands. Norsk Polarinstitutt, medd. nr. 133, Oslo, 57 pp. Fjellberg, A., 1998. The Collembola of Fennoscandia and Denmark. I: Poduromorpha and Onychiuridae. Fauna Entomologica Scandinavica 35. EJ Brill, The Netherlands, 184 pp. Frampton, G.K., Van den Brink, P.J., Wratten, S.D., 2001. Diel activity patterns in an arable collembolan community. Appl. Soil Ecol. 17, 63–80. Franchini, P., Rockett, C.L., 1996. Oribatid mites as ‘‘indicator’’ species for estimating the environmental impact of conventional and conservation tilling practices. Pedobiologia 40, 217–225. Gjelstrup, P., Søchting, U., 1984. Oribatid mites (Acarina) dominant in some lichen and moss species of maritime rocks on Bornholm in the Baltic. In: Griffiths, D.A., Bowman, C.E. (Eds.), Acarology VI, vol. 1, John Wiley & Sons Ltd., Chichester, pp. 528– 533. Gonzalez, A., Chaneton, E.J., 2002. Heterotroph species extinction, abundance and biomass dynamics in an experimentally fragmented microecosystem. J. Anim. Ecol. 71, 594–602. Grubb, P.J., 1987. Some generalizing ideas about colonization and succession in green plants and fungi. In: Gray, A.J., Crawley, M.J., Edwards, P.J. (Eds.), Colonization, Succession and Stability. Blackwell Scientific Publications, Oxford, pp. 81–102. Hale, W.G., 1965. Observations on the breeding biology of Collembola. Pedobiologia 5, 146–152. Hammer, M., 1972. Microhabitats of oribatid mites on a Danish woodland floor. Pedobiologia 12, 412–423. Hansson, A.-C., Andre´ n, O., Bostro¨ m, S., Bostro¨ m, U., Clarholm, M., Lagerlo¨ f, J., Lindberg, T., Paustian, K., Pettersson, R., Sohlenius, B., 1990. Structure of the agroecosystem. In: Andre´ n, O., Lindberg, T., Paustian, K., Rosswall, T. (Eds.), Ecology of Arable Land – Organisms, Carbon and Nitrogen Cycling. Ecol. Bull. 40, Copenhagen, pp. 41–83. Hopkin, S., 1997. Biology of the Springtails (Insecta: Collembola). Oxford University Press, Oxford, 330 pp. Hoy, J.B., 1990. Long-term effects of lindane on the oribatid community in a pine plantation. Environ. Entomol. 19, 898–900. Hu¨ lsmann, A., Wolters, V., 1998. The effects of different tillage practices on soil mites, with particular reference to Oribatida. Appl. Soil Ecol. 9, 327–332. Hutson, B.R., 1980. Colonization of industrial reclamation sites by acari, collembola and other invertebrates. J. Appl. Ecol. 17, 255– 275. Joosse, E.N.G., 1969. Population structure of some surface dwelling collembola in a coniferous forest soil. Neth. J. Zool. 19, 621– 634. Koehler, H., 1998. Secondary succession of soil mesofauna: A thirteen year study. Appl. Soil Ecol. 9, 81–86. Lindberg, N., 2003. Soil Fauna and Global Change—Responses to Experimental Drought, Irrigation, Fertilisation and Soil Warming. PhD thesis, Silvestria 270, Swedish University of Agricultural Sciences, Uppsala. Lindberg, N., Bengtsson, J., Persson, T., 2002. Effects of experimental irrigation and drought on the composition and diversity of soil fauna in a coniferous stand. J. Appl. Ecol. 39, 924–936.
174
N. Lindberg, J. Bengtsson / Applied Soil Ecology 28 (2005) 163–174
Loreau, M., Mouquet, N., Gonzalez, A., 2003. Biodiversity as spatial insurance in heterogeneous landscapes. Proc. Natl. Acad. Sci. U.S.A. 100, 12765–12770. Lucarotti, C.J., 1981. The effect of fire and forest regeneration on mesofauna populations and microfungal species in lichen woodland soils. McGill Subarctic Res. Paper 32, 7–26. Luxton, M., 1981. Studies on the oribatid mites of a Danish beech wood soil. IV. Developmental biology. Pedobiologia 21, 312– 340. Luxton, M., 1982. The ecology of some soil mites from coal shale tips. J. Appl. Ecol. 19, 427–442. MacArthur, R.H., Wilson, E.O., 1967. The Theory of Island Biogeography. Princeton University Press, New Jersey. Maraun, M., Scheu, S., 2000. The structure of oribatid mite communities (Acari, Oribatida): patterns, mechanisms and implications for future research. Ecography 23, 374–383. Mebes, K.-H., Filser, J., 1997. A method for estimating the significance of surface dispersal for population fluctuations of Collembola in arable land. Pedobiologia 41, 115–122. Niemi, R., Karppinen, E., Uusitalo, M., 1997. Catalogue of the Oribatida (Acari) of Finland. Acta Zool. Fenn. 207, 1–39. Nilsson, L.-O., 1997. Manipulation of conventional forest management practices to increase forest growth – results from the Skogaby project. For. Ecol. Manage. 91, 53–60. Norton, R.A., 1994. Evolutionary aspects of oribatid mite life histories and consequences for the origin of the Astigmata. In: Houck, M. (Ed.), Mites. Ecological and Evolutionary Analyses of Life-History Patterns. Chapman & Hall, New York, pp. 99–135. Norton, R.A., Kethley, J.B., Johnston, D.E., O’Connor, B.M., 1993. Phylogenetic perspectives on genetic systems and reproductive modes of mites. In: Wrensch, D.A., Ebbert, M.A. (Eds.), Evolution and Diversity of Sex Ratio in Insects and Mites. Chapman & Hall, New York, pp. 8–99. Ojala, R., Huhta, V., 2001. Dispersal of microarthropods in forest soil. Pedobiologia 45, 443–450. Persson, T., Lohm, U., 1977. Energetical significance of the annelids and arthropods in a Swedish grassland soil. Ecol. Bull., 23, Swedish Natural Science Research Council, Stockholm, pp. 56– 141. Petersen, H., 1995. Temporal and spatial dynamics of soil Collembola during secondary succession in Danish heathland. Acta Zool. Fenn. 196, 190–194. Ronce, O., Perret, F., Olivieri, I., 2000. Landscape dynamics and evolution of colonizer syndromes: interactions between reproductive effort and dispersal in a metapopulation. Evol. Ecol. 14, 233–260. SAS Institute, 2000. SAS/Stat User’s guide, Version 8. SAS Institute Inc., Cary, North Carolina. Schaefer, M., Schauermann, J., 1990. The soil fauna of beech forests: comparison between a mull and a moder soil. Pedobiologia 34, 299–314. Shaw, P.J.A., 1997. Post-fire successions of Collembola in lowland heaths in South-Eastern UK. Pedobiologia 41, 80–87.
Siepel, H., 1995. Applications of microarthropod life-history tactics in nature management and ecotoxicology. Biol. Fert. Soils. 19, 75–83. Siepel, H., 1996. Biodiversity of soil microarthropods: the filtering of species. Biodivers. Conserv. 5, 251–260. Simberloff, D., 1981. What makes a good island colonist? In: Denno, R.F., Dingle, H. (Eds.), Insect Life History Patterns. Springer, New York, pp. 195–205. Skubala, P., 1995. Moss mites (Acarina: Oribatida) on industrial dumps of different ages. Pedobiologia 39, 170–184. SMHI, 1998. Va¨ dera˚ ret 1998. Va¨ der och Vatten. SMHI, Norrko¨ ping, 8 pp. (in Swedish). Smrz, J., Kocourkova, J., 1999. Mite communities of two epiphytic lichen species (Hypogymnia physodes and Parmelia sulcata) in the Czech Republic. Pedobiologia 43, 385–390. Stearns, S.C., 1992. The Evolution of Life Histories, Oxford University Press, Oxford. Steiner, W.A., 1994. The influence of air pollution on moss-dwelling animals: 1. Methodology and composition of flora and fauna. Rev. Suisse Zool. 101, 533–556. Steiner, W.A., 1995. Influence of air pollution on moss-dwelling animals: 3. Terrestrial fauna, with emphasis on Oribatida and Collembola. Acarologia 36, 149–173. St. John, M.G., Bagatto, G., Behan-Pelletier, V., Lindquist, E.E., Shorthouse, J.D., Smith, I.M., 2002. Mite (Acari) colonization of vegetated mine tailings near Sudbury, Ontario. Canada Plant Soil. 245, 295–305. Strojan, C.L., 1978. The impact of zinc smelter emissions on forest litter arthropods. Oikos 31, 41–46. Tamm, J.C., 1984. Surviving long submergence in the egg stage – a successful strategy of terrestrial arthropods living on flood plains (Collembola, Acari, Diptera). Oecologia 61, 417–419. Tamm, J.C., 1986. Fu¨ nfja¨ hrige Collembolensukzession auf einem verbrannten Kiefern-waldboden in Niedersachsen (BRD). Pedobiologia 29, 113–127. Verhoef, H.A., Van Selm, A.J., 1983. Distribution and population dynamics of Collembola in relation to soil moisture. Holarctic Ecol. 6, 387–394. Walker, B., 1995. Conserving biological diversity through ecosystem resilience. Conserv. Biol. 9, 747–752. Wanner, M., Dunger, W., 2002. Primary immigration and succession of soil organisms on reclaimed opencast coal mining areas in Eastern Germany. Eur. J. Soil Biol. 38, 137–143. Webb, N.R., 1994. Post-fire succession of cryptostigmatid mites (Acari, Cryptostigmata) in a Calluna-heathland soil. Pedobiologia 38, 138–145. Weigmann, V.G., Kratz, W., 1981. Die deutschen Hornmilbenarten und ihre o¨ kologische Charakteristik. Zoologische Beitra¨ ge. 27, 459–489. Whelan, J., 1978. Acarine succession in grassland on cutaway raised bog. Sci. Proc. R. Dublin Soc. 6, 175–183. Winkler, H., Kampichler, C., 2000. Local and regional species richness in communities of surface-dwelling grassland Collembola: indication of species saturation. Ecography 23, 385–392.