The importance of litter for
interactions between terrestrial plants and invertebrates
Maria Gelfgren
Student
Degree Thesis in Biology 30 ECTS Master’s Level
Report passed: 3 March 2010 Supervisor: Michael Schneider
1
Abstract
According to the exploitation ecosystem hypothesis (EEH), terrestrial ecosystems are characterized by well defined trophic levels and strong trophic interactions with community level tropic cascades. In unproductive terrestrial habitats as tundra heaths, the energy shunt from litter and apparent competition between herbivores and detritivores are expected to be important for the structure and dynamics of the invertebrate community. The aim of this study was to test this hypothesis by investigating if plant litter accumulation was affecting the invertebrate community on a nutrient-poor tundra heath. The study was performed during one summer on the highland part of Joatka research area, in the north of Norway.
The experimental area included 16 plots (100 m2 each), of which 12 had been litter-
manipulated. On four plots the amount of litter was increased by 100 %, on four by 200 % and on four by 400 %. Four plots were untreated and used as control plots. Invertebrates were collected by emergence traps (which cover an area of 1 m2), one trap on each plot and one pitfall trap inside each emergence trap. During the study period, traps were emptied and moved twice, resulting in three sampling periods. The invertebrates collected were counted and their length was measured, than all invertebrates were sorted into taxa and trophic guilds.
During the study period, herbivore grazing damage was investigated on all 16 experimental plots, signs of herbivores on leaves of vascular plants in an area covering 3 m2 per plot were noted, for every leaf with signs of herbivory the percentage of leaf area removed was
estimated.
Plant biomass and plant species composition were estimated in all experimental plots by harvesting above-ground plant parts. In each plot, two squares were randomly chosen and all biomass in this square was collected. Plant biomass was sorted in to following groups: dwarf birch, billberry, Salix herbacea, Salix spp, graminoids, herbs, lichens, mosses and dwarf shrub. Before weighing the plant material, it was stored in paper bags at room temperature and then dried for 48 h at 40°C. In order to detect fertilisation effects, all bilberry shoots that had been produced during the actual summer were separately weighted when analyzing the plant biomass.
The result showed that the invertebrate community in this area is dominated by carnivores while detritivores, parasitoids and herbivores are quite rare, this was in accordance with previous studies made in the area. Litter manipulation did not create any significant variation in the community structure, but there was a slight tendency that carnivore biomass increased and biomass of herbivores decreased when litter was added to the system. In contrary to this, gracing activity especially on dwarf willow (Salix herbacea) increased in plots were 100 % and 200 % more litter was added. There is a positive correlation between biomass of herbivores and detritivores but the reason for this seems unclear. No fertilisation effect was detected in litter manipulated plots. The structure and dynamics of the actual community could not be described by the food web theory EEH and energy shunt from litter and apparent competition between herbivores and detritivores. It seems to be several complicating factors to take into consideration when describing this community.
2
Contents
Introduction ...3-5 Material and methods ...6-9 Results ...10-18 Discussion ...18-22 References ...22-24 Acknowledgements ... 24
3
Introduction
What creates the population structure and dynamics in natural communities? These topics have been discussed vigorously by ecologists for a long time (Aunapuu et al. 2008, Fretwell 1987, Hairston et al. 1960, Holt and Polis 1997, Hunter and Price 1992, Oksanen and
Oksanen 2000, Oksanen et al. 1981, Polis 1994, Polis 1999, Polis et al.1998, Polis and Strong 1996, Strong 1992, White 1978). Food webs with well defined trophic levels of top predators, intermediate species, and basal species, have been used in the history of ecology to describe population structure and dynamics (Pimm et al. 1991). Top - down (consumption) effects that create trophic cascades through the web (Chase 1998, Strong 1992, Marquis and Whelan 1994) and bottom - up (resource) effects have been regarded as the main influence forces (Hunter and Price 1992, Menge 1992, Persson 1999, Power 1992, Strong 1992).
One food web theory assumes that bottom-up forces are most important and that green plants determine the population size of herbivores, which limits carnivore populations (White 1978).
According to these theories, dead plant material accumulates in the litter layer because plants have too low nutrient content or different types of defence mechanisms, which makes them unsuitable to their consumers (White 1978).
Ecologists behind “The world is green hypothesis” (HSS) (Hairston et al. 1960) and the Hypothesis of exploitation ecosystems (EEH), (Aunapuu et al. 2008, Oksanen et al. 1981, Oksanen and Oksanen 2000) regard top-down forces as most important. Carnivores are assumed to regulate herbivore density, and in this way release plants from grazing pressure and allow plant biomass to accumulate in the litter layer (Aunapuu et al. 2008, Marquis and Whelan 1994, Oksanen et al. 1981, Oksanen and Oksanen 2000). According to the EEH hypothesis, the number of trophic levels is connected to the primary productivity and
carnivores can only be present in productive habitats, while less productive habitats consist of plants and herbivores or only by plants in very barren habitats (Aunapuu et al. 2008, Oksanen et al. 1981, Oksanen and Oksanen 2000).
This type of food web theories have been used to describe the dynamic in terrestrial habitats with endotherms and plants (Aunapuu et al. 2008, Oksanen et al. 1981, Oksanen and Oksanen 2000, Rammul et al. 2007). Critics argue that these theories are too simple to be descriptive and predictive of nature, that distinctive trophic levels do not exist and trophic cascades are only restricted to few and special cases in nature (Polis 1991, 1994, 1999). Further critical argumentation is that invertebrates have not been correctly included in terrestrial food webs and because of their species richness they are believed to be central in the structure of
terrestrial communities (Polis 1991, 1994, 1999). In tundra sites biomass of invertebrates has been found to be larger than biomass of vertebrates (MacLean 1981).
Complicating factors when including invertebrates in food web theories are for example that invertebrates can be omnivores (Pimm et al. 1991) and have ontogenetic nich shifts (Speight et al. 1999), due to this, trophic level are difficult to decide (Power 1992, Schneider and Grellman 2000). Carnivore populations are size structured and intra guild predation, IGP, might occur, this makes it difficult to interpret their impact on their resource and their own population (Holt and Polis 1997, Polis and Holt 1992, Polis et al. 1989). Omnivory blurs distinct trophic levels and hence trophic cascades, and this makes food chain theory hypothesis less testable (Diehl 1993, Polis and Strong 1995).
4
In several terrestrial communities dead plant material accumulates in the litter layer together with animal tissues and waste products (Speight et al. 1999). Litter creates the basic resource for the detritivorous part of the food web and are consumed by many different species of invertebrates, fungi and bacteria (Speight et al. 1999). The importance of the litter layer for community structure and dynamic is poorly understood and energy shunts from detritus may have important effects (Polis 1991, Polis and Strong 1996).
In litter-rich systems as the Norwegian tundra, insectivorous carnivores might be able to build up dense equilibrium populations due to a strong energy shunt from litter via detritivores (Oksanen et al 1997, Schneider and Grellman 2000). If carnivores utilise herbivores as well as detritivores, apparent competition between these groups could reduce herbivore equilibrium population density. This may affect herbivore-plant interactions, and the food chain based on living plants might not be dynamically important in such a system (Oksanen et al. 1997, Schneider and Grellman 2000).
Norwegian tundra
Oksanen et al. (1997) studied composition of the invertebrate community on the Norwegian tundra and according to their results, litter seems to be important because invertebrate communities are characterised by a large amount of carnivores while herbivores are almost absent.
Schneider and Grellman (2000) made a fertilisation experiment on a Norwegian tundra heath, in which they studied trophic level composition of invertebrates and herbivore impact on plants. Their results showed that fertilisation leads to an increasing amount of litter in the system and more litter leads to increasing density of carnivores (Schneider and Grellman 2000). Herbivore density did not change with an increasing amount of litter but grazing activity decreased, and detritivore density did not respond to litter accumulation (Schneider and Grellman 2000). The results of their study were similar to those of the study made by Oksanen et al. (1997), i.e. carnivores are dominating the invertebrate community of the tundra and the energy shunt from litter via detritivores seems to be important (Schneider and
Grellman 2000).
The aim of this study was to examine the effect of litter accumulation on the invertebrate community on a Norwegian tundra heath (Figure 1). Fertilisation leads to change in plant community (Grellman 2001, Ljungberg 2002), and this probably affected the quality of the litter and the result of the study made by Schneider and Grellman (2000). In the present study, plant litter manipulation were done by collecting heath-specific plant litter and add this litter to undisturbed plots of heath vegetation, two years before the actual study started.
If there are strong trophic interactions, apparent competition between detritivores and herbivores and plant litter are important in this system, the predictions according to EEH are that an increasing mass of litter should cause: 1) an increased equilibrium density of
carnivores, 2) a decreased equilibrium density of herbivores and due to this, diminishing grazing effects on plants, and 3) an unchanged detritivore population.
5
Apparent Competition
Birds Parasitoids
Carnivores
Detritivores Herbivores
Plants Litter
Figure 1. A likely invertebrate food web on the tundra heath redrawn from Schneider and Grellman (2000).
The thickness of the arrows indicates the suggested importance of the pathway in the web and the bold borders shows the compartments which are supposed to be dynamically important.
6
Material and methods
The fieldwork started on 20 June 2000, when snow covered large parts of the study area and ended on 1 September 2000, when the study area started to turn into autumn colours.
Studied area
The study was performed on the highland part of the Joatka research area (at 545 m above sea level), a tundra system on the Finmarksvidda plateau in the north of Norway (69°46’N,
23°57’E) (Figure 2). In the highland, snow usually disappears in late June and returns in late September. In late March or early April the greatest thickness of the snow layer is reached, which is about 100 cm. The growing season lasts for two months and during this short summer the sun is shining 24 hours per day. Mean daily temperature differs greatly within a range from about +5 to +20°C. Annual precipitation is relatively low and averages 364 mm, most of it coming as rain during summer (Schneider and Grellman 2000). Studies of
interactions between plants, microtine rodents and predators have been conducted in the Joatka research area since 1977 (Aunapuu et al. 2008, Moen and Oksanen 1991, Oksanen and Oksanen 2000, Oksanen et al. 1981, 1997, Turchin et al. 2000).
Studied system
The plant community at the study sight is a dwarf shrub heath dominated by dwarf birch (Betula nana), dwarf willow (Salix herbacea), bilberry (Vaccinum myrtillus) and crow berry (Empetrum nigrum ssp hermaphroditum). In the study area Norwegian lemmings (Lemmus lemmus) and grey-sided voles (Clethrionomys rufocanus) are common herbivorous rodents.
Reindeer (Rangifer tarandus) pass trough the area on their seasonal migrations in spring and autumn. Long-tailed jaegers (Stercorarius longicaudus), plovers (Charadrius morinellus, Pluvaris apricaria), wheatears (Oenanthe oenanthe) and medow pipits (Anthus pratensis) are the most common vertebrate insectivores. Ptarmigan (Lagupus mutus) are less common and shrews (Sorex spp) are very rare in this vegetation type. The invertebrate community are dominated by carnivores and the most common carnivore taxa are Araneae, Carabidae, Opiliones and Staphylinidae. Herbivores consist of Byrrhidae (moss eaters), different species of Lepidoptera, Coleoptera and Hymenoptera (leaf chewers), Cecidomyiidae (gall builders), Aphidae (sapsuckers), Chloropidae and Syrphidae (miners). Detrivores include Acarina, Collembola, Silphidae, Psocoptera and 17 families of Diptera. The most common parasitoids belong to the hymenopteran taxa Proctotrupoidea and Ichneumonidae (Schneider and
Grellman 2000).
Experimental set-up
The experimental area was composed of 16 plots (100 m2 each), of which 12 plots had been litter-manipulated (litter had been added) during the summer of 1998 (Figure 2). On four plots the amount of litter was increased by 100 %, on four by 200 % and on four by 400 % (Figure 2). Four plots were used as control and in those plots no litter was added (Figure 2).
7
A:
0 % 400 % 100 % 200 %
B:
200 % 0 % 400 % 100 %
C:
100 % 200 % 0 % 400 % 10 m
D:
400 % 100 % 200 % 0 %
Figure 2. Map of Joatka research area (above) and experimental plot (below).
Alta
8
Invertebrates
Emergence traps were used to collect invertebrates (Figure 3). These traps cover an area of 1 m2 each and consist of a basal metal ring, a funnel tent and a collecting jar at the top. The basal metal ring was dug into the ground to prevent animals from escaping the area. The funnel tent was attached to the outside of the metal ring in order to direct invertebrates to the collecting jar on the top of the tent (ca 1.5 m above ground) were they were killed and preserved in glycol. Inside these emergence traps, close to the wall of the metal ring, a pitfall trap with glycol was placed to sample invertebrates from the ground. This type of traps makes it possible to calculate the number of invertebrates caught per unit area (1 m2). During the actual summer, traps were emptied and moved twice within the experimental plot, resulting in three sampling periods of three weeks each. Invertebrates were first sorted into taxa (Chinery 1993, Douwes et al. 1997, Hedström 1995, Meinander and Panelius 1969), and then into trophic guilds (Schneider and Grellman 2000). For those taxa where adults and larvae are feeding on the same resource, the guild is easy to decide. In taxa were adults and immatures feed on different resources, the life stage that gains most biomass is the most important for this study and this determines their trophic position (Schneider and Grellman 2000). The same trophic guilds were used as in Schneider and Grellman (2000). All invertebrates were counted and their length was measured, then length-weight regression models from the literature were used to compute the biomass of all invertebrate taxa (Breymeyer 1967, Henschel et al. 1996, Rogers et al. 1976, 1977).
Collecting jar, cross-section:
Plastic tube
Basal metal ring Collecting jar
Glycol
Cover
Funnel tent
Anchorage
Figure 3. A picture and an outline of one emergence trap with the collecting jar in cross section.
9
Herbivore grazing
In the middle of August herbivore grazing damage was measured in all 16 experimental plots. Signs of herbivore damage on leaves of all vascular plants present within three transects on each square were noted. Each transect was 10 m long and 10 cm wide, thus in total covering 3 m2 per plot. The number of leaves damaged by herbivorous invertebrates was counted for all vascular plant species along each transect. The percentage of leaf area removed was estimated for every leaf with signs of herbivory. Herbivore damage in plots with different manipulations was compared using an ANOVA on ln transformed data.
Plant biomass
Plant species composition and plant biomass in the experimental plots were estimated by harvesting above-ground plant parts at the time of maximum plant biomass, in early August. In each of the 16 plots, two squares (25 cm×25 cm) were randomly chosen and all plant biomass was collected. The plant material was assigned to the following groups:
dwarf birch, billberry, Salix herbacea, Salix spp, graminoids (eg. Deschampsia flexuosa, Festuca ovina, Carex bigelowii, Juncus trifidus), herbs (Pedicularis lapponica, Trientalis europaea), lichens, mosses, dwarf shrub (Empetrum nigrum ssp. hermaphroditum, Vaccinum uliginosum, Vaccinum vitis-idaea, Phyllodoce caerulea, Andromeda polifolia).
The plant material was stored in paper bags for several months at room temperature and then dried for 48 h at 40°C before weighing.
Fertilisation effect
In order to detect possible fertilisation effects of the litter manipulation, all bilberry shoots that have been produced during the year were separately treated when analyzing the plant biomass.
Meteorological data
Data on precipitation were collected at Joatka fjellstue, which is situated ca 1 km south of the study area (Figure 2). Temperature measurements were taken at Suolovuompi fjellstue 20 km south of the experimental area and compiled by the Norwegian Meteorological Institute. In order to analyse the meteorological data, the summer was subdivided into five two-week periods as in the study of Schneider and Grellman (2000). The mean temperature under one period was the average of all daily mean temperatures. The precipitation was the sum of all precipitation during one period.
10
Results
Total biomass of invertebrates was 16951.75 mg. All invertebrate guilds, taxa, total number of each taxa, dry weight and percentage of dry weight (actual guild and total biomass) are shown in Table 1.
Table 1. Trophic guilds, taxa, total dry weight (mg), total number and percentage dry weight of the actual guild and all collected invertebrates.
Carnivores Total dry weight
(mg)
Number Percentage
% Fam. Cantharidae
Fam. Carabidae Fam. Dolichopodidae
Fam. Empididae Fam. Staphylinidae
Ord. Opiliones Ord. Aranea Coleoptera larvae
Total:
92.40 1322.33
0.99 56.28 8073.87
864.44 2369.11
114.48 12893.91
10 125
1 87 3104
60 2285
22 5694
0.72 10.26 7.68·10-3
0.44 62.62
6.70 18.37
0.89 76.07 Detritivores Total dry weight
(mg)
Number Percentage
% Fam. Anthomyiidae
Fam. Bibionidae Fam. Chironomidae
Fam. Calliphoridae Fam. Carnidae Fam. Muscidae Fam. Mycetophilidae
Fam. Phoridae Fam. Sciaridae Fam. Scathophagidae
Fam. Tipulidae Ord. Acarina Ord. Collembola
Diptera larvae Total:
10.57 61.15 54.95 33.91 1.53 241.40
59.74 142.50 222.11 0.99 98.79 30.50 513.43
0.04 1471.61
8 8 240
16 8 255
46 638 967 1 2 698 5227
1 8115
0.72 4.16 3.73 2.30 0.10 16.40
4.06 9.68 15.09 67.27·10-3
6.71 2.07 34.89 2.72·10-3
8.68 Parasitoids Total dry weight
(mg)
Number Percentage
% Fam. Tachinidae
Subord. Apocrita Total:
12.56 192.63 205.19
7 737 744
6.12 93.88
1.21
Others Total dry weight
(mg)
Number Percentage
% Fam. Culicidae
Fam. Simuliidae/
Fam.Ceratopogonidae Ord. Plecoptera Ord. Trichoptera Ord. Ephemeroptera
Total:
156.43 53.99 71.48 11.88 4.99 298.77
76 72 11 5 1 165
51.69 18.07 23.93 3.98 1.67 1.76
11
Herbivores Total dry weight
(mg)
Number Percentage
% Fam. Boreidae
Fam. Byrrhidae Fam. Cecidomyiidae
Fam. Chloropidae Fam. Chrysomelidae
Fam. Curculionidae Fam. Syrphidae Subord. Symphyta
Ord. Homoptera Ord. Lepidoptera Ord. Thysanoptera Lepidoptera larvae Symphyta larvae
Total:
10.42 644.93
18.00 1.37 269.12
66.14 19.83 19,52 9.95 529.32
0.19 178.23 315.25 2082.27
2 35 133
3 66
7 8 13 42 32 1 27 13 382
0.05 30.97
0.09 0.01 12.92
3.18 0.1 0.09 0.05 25.42 9.12·10-3
8.53 15.13 12.28
Biomass (dry weight mg /m2) of all feeding guilds in the different treatments are shown in Figure 4.
0 200 400 600 800 1000 1200
Carnivores Detritivores Herbivores Others Parasitoids Biomass (dry weight mg / m2 )
0 100 200 400
Figure 4. Biomass (mg dry weight / m2) of invertebrates from different feeding guilds in the different treated plots. 0 (control plots), 100 (100 % more litter), 200 (200% more litter) and 400 (400 % more litter).
12
Biomass (dry weight mg /m2) of moss eating invertebrates in different treatments is showed in Figure 5.
0 20 40 60 80 100 120 140
0 100 200 400
Biomass (dry weight mg / m2 )
Figure 5. Biomass (mg dry weight / m2) of moss eating invertebrates in the different treated plots. 0 (control plots), 100 (100 % more litter), 200 (200% more litter) and 400 (400 % more litter)
Carnivores
Total carnivore biomass was 12893.91 mg, the total number was 5694 and 76.07 % of all trapped invertebrates were carnivores (Table 1). The most common carnivore taxa were Staphylinidae 62.62 %, Aranea 18.37 % and, Carabidae 10.26 % (Table 1). According to the result, there is a tendency that carnivore biomass increases with increasing amount of litter but differences between treatments are not significant (Figure 4).
Detritivores
Total biomass of detritivores was 1471.61 mg, total number 8115 and of all invertebrates 8.68
% were detritivores (Table 1). The most common taxa were Collembola 34.89 %, Muscidae 16.40 % and Sciaridae 15.09 % (Table 1). Biomass of detritivores in manipulated plots was not significantly higher than in control plots. The tendency is that detritivore biomass
increases when adding 100 % more litter but there is no response of change in biomass when adding 200 % and 400% (Figure 4).
13
Parasitoids
The total number of parasitoids was 744 and total biomass was 205.19 mg, only 1.21 % of the collected invertebrates were parasitoids (Table 1). The most common parasitoids were
parasitic wasps (Proctotrupoidea and Ichneumonidae) 93.88 %. Adding litter to the system affected biomass of parasitoids negatively, but there is no significant difference between manipulated plots and control plots (Figure 4).
Herbivores
The most common herbivores were moss eating taxa Byrrhidae 30.97 %, leaf chewing insects as Lepidoptera (larvae 8.53 % and adults 25.42 %) and Chrysomelidae 12.92 % (Table 1).
Total herbivore biomass was 2082.27 mg, total number 382 and 12.28 % of the total mass of invertebrates were herbivores (Table 1). Litter manipulation did not show any significant effect on herbivore biomass, but the tendency was that litter in the system leads to decreasing biomass of herbivore (Figure 4). Moss eating taxa (Byrrhidae and Boreidae) were most common in untreated plots (Figure 5).
There is a positive correlation between herbivore biomass and detritivore biomass (Figure 6).
R² = 0,356 P = 0,015
0 50 100 150 200 250
0 50 100 150 200 250 300
Herbivore biomass (mg/m2)
Detritivore biomass (mg/m2)
Figure 6. The correlation between biomass (mg/m2) of detritivores and herbivores in the study area. Each dot represents one experimental plot.
14
Others
The invertebrate taxa belonging to this group are shown in Table 1, total biomass was 298.77 mg, and total number 165 and 1.76 % of the total mass were others. Culicidae, 51.69 % is the most common family in this group. This group of invertebrates did not response to litter manipulations (Figure 4).
Herbivory
Dwarf willow, Salix herbacea, is one of the most palatable plant species in this area (Aunapuu et al. 2008) and herbivores seemed to graze more on dwarf willow in manipulated plots than in control plots, especially on plots treated with 100 % and 200 % more litter (Table 2, Figure 7 and 8). But there was no significant difference in herbivore activity between treatments.
Herbivore activity on Vaccinum myrtillus and Betula nana was almost the same in all treatments (Figure 7 and 8). The tendency is the same when grazing damage on all plant species is calculated, herbivore activity is higher in plots manipulated with 100 % and 200 % more litter but in plots treated with 400 % more litter grazing activity is equal to control plots (Table 2, Figure 9 and 10). Damage by chewing herbivores was most common while miners were rare and no signs of galls were detected.
Table 2. The results of an ANOVA on ln-transformed data on herbivory.
Variable df Effect MS effect df Error MS Error F P Number of leaves 3 0.058 12 0.018 3.307 0.057
Leaf area index 3 0.072 12 0.012 6.185 0.009
15 0
50 100 150 200 250 300 350 400 450 500
Salix herbacea Vaccinium myrtillus Betula nana Average total area removed by herbivores (mm2 / m2)
0 100 200 400
Figure 7. Average total area (mm2 / m2) removed by herbivores from Salix herbacea, Vaccinium myrtillus and Betula nana in the different treated plots: 0 (control plots), 100 (100 % more litter), 200 (200% more litter) and 400 (400 % more litter).
0 5 10 15 20 25
0 100 200 400
Mean no. of leaves damaged / m²
Amount of litter added
Salix herbacea Vaccinium myrtillus Betula nana
Figure 8. Mean number of leaves damaged / m2 on Salix herbacea, Vaccinium myrtillus and Betula nana in the different treated plots: 0 (control plots), 100 (100 % more litter), 200 (200% more litter) and 400 (400 % more litter)
16 0
200 400 600 800 1000 1200
0 100 200 400
Average total area removed by herbivores mm2 / m2
Amount of litter added
Figure 9. Average total area (mm2 / m2) removed by herbivores from all plant species in the different treated plots: 0% (control plots), 100 (100 % more litter), 200 (200% more litter) and 400 (400 % more litter).
0 10 20 30 40 50 60
0 100 200 400
Mean no. of leaves damaged / m²
Amount of litter added
Figure 10. Mean number of leaves damaged / m2on all plant species in the different treated plots: 0 (control plots), 100 (100 % more litter), 200 (200% more litter) and 400 (400 % more litter).
17
Plant biomass
Litter manipulation had no effect on plant biomass and species composition, i.e. no fertilisation effect was detected (Table 3).
Table 3. The mean and standard error (SE) of the dry weight (mg/m2) of plant biomass in the different treated experimental plots.
Mean
0 % 100 % 200 % 400 %
Bilberry, leaf 35 35.5 32.8 31
Billberry, steam/twig 71.4 85.8 70.6 71
Bilberry, year shoot 11.4 15.9 11.8 12
Dwarf birch, leaf 23 13.6 25.8 18
Dwarf birch, steam/twig 100 124 161 101
Salix herbacea, leaf 0.94 0.88 0.76 0.5
Salix herbacea, steam/twig 0.74 0.8 0.6 3.9
Graminoids 11.2 2.4 2.56 6.1
Lichens 2.88 4.18 3.14 2.7
Mosses 4.04 2.06 3.98 1.4
Dwarf shrub 147 144 110 189
Herbs 2.6 10.1 12.8 5.8
SE
0 % 100 % 200 % 400 %
Bilberry, leaf 6.99 2.16 5.9 8.2
Billberry, steam/twig 17 12.5 17 17
Bilberry, year shoot 1.71 3.49 2,5 3.2
Dwarf birch, leaf 13.1 8.76 6 5.5
Dwarf birch, steam/twig 54.6 76.4 43 36
Salix herbacea, leaf 0.25 0.09 0.5 0.3
Salix herbacea, steam/twig 0.37 0.3 0.3 3.5
Graminoids 6.22 0.88 1.5 3
Lichens 0.6 2.38 0.9 1.2
Mosses 2.85 0.44 2.3 0.5
Dwarf shrub 22.7 35.2 25 13
Herbs 1.24 4.79 3.4 1.4
18
Meteorological data
During the period of fieldwork, the highest mean temperature 19.6 °C was reached in the middle of July. The period 16-31 July was warmest. Precipitation was highest during the first part of the period, 16-30 June, with maximum daily precipitation 17.9 mm/m2 (Figure 11).
0 5 10 15 20 25 30 35 40 45 50
0 2 4 6 8 10 12 14 16
16-30 June 01-15 July 16-31 July 01-15 August 16-31 August Precipitation (mm/m2)
Mean temperature (°C)
mean temperature precipitation
Figure 11. The pooled precipitation (mm/m2) and mean temperature (C°) during the period of field study.
Discussion
Litter manipulation on the actual tundra heath did not create any significant changes in the structure of the invertebrate community (Figure 4). In all different treated plots the
community structure was the same, carnivore biomass largely exceeded the biomass of herbivores and detritivores while parasitoids were very rare in the system (Figure 4, Table 1).
This pattern was in accordance with earlier studies made in the same area (Oksanen et al.
1997, Schneider and Grellman 2000) and indicates that this structure is stable and there are probably several factors responsible for it, not only energy shunt from litter (Schneider and Grellman 2000). Oksanen et al. (1997) concluded that invertebrate herbivores were almost
19
absent on this tundra heath, in their study sap and mould feeding insects were classified as others. In the present study and the study made by Schneider and Grellman (2000), sap and mould feeders were classified as herbivores and detritivores, respectively. One problem with studies on taxonomic groups is the limited knowledge about invertebrate biology (MacLean 1981). Oksanen et al. (1997) might have classified herbivores as members of detritivore, carnivore or parasitoid guilds and this can explain why herbivores seemed to be absent. In this study invertebrates belonging to fam. Tipulidae are classified as detritivores, they are
rhizophages and feeds on dead and living plant roots, the right classification might have been herbivores (Chernov 1985).
In the actual study, biomass of herbivores exceeded biomass of detritivores (Table 1), in the study made by Schneider and Grellman (2000) the result showed the opposite pattern. As mentioned in the introduction part, plant community, plant quality, litter quality and plant structure changed due to fertilisation in the study made by Schneider and Grellman (2000), this could have effects on the invertebrate community composition (Grellman 2001, Ljungberg 2002). Vascular plants are favoured by fertiliser and this could favour some detritivores, rhizophagous Tipulidae as an example (Cernov 1985). Fertilising the tundra, affects lichen cover negatively (Grellman 2001), and lichens are important for collembolans, both as habitat and food (Chernov 1985). Thus, fertiliser could also have negative influence on the density of detritivores (Chernov 1985). Some herbivores can respond positively to plant quality change, aphids as an example, have the ability to gain from nitrogen fertilisation (Kytö et al. 1996) and in the study conducted by Schneider and Grellman (2000) the number of aphids in 1999 was 1646 while in the present study homopteran insects were rare (42 animals) and constituted only 0.5 % of all herbivores. Herbivores can also respond negatively to fertilisation factors (Kytö et al. 1996), fertilisation has, for example, negative effects on the abundance of mosses (Jonasson et al. 1999), and this might have effected biomass of moss eating herbivores negatively in the study made by Schneider and Grellman (2000).
Total biomass of invertebrates at the actual tundra site was higher in 1998, ≈ 32000 mg / 16 m2 and 1999, ≈ 26000 mg / 16 m2 (Schneider and Grellman 2000), than in this study,
16951.75 mg / 16 m2. Birds were excluded from some of the experimental plots in 1998 and 1999, in order to track effects of bird predation (Schneider and Grellman 2000). It is likely that bird exclusion has a positive effect on total invertebrate biomass. During fieldwork in the present study, funnel tents of some traps came off and due to this invertebrate were able to escape. It was a few occasional but this might affect total biomass of invertebrates negatively.
Effects of bird exclusion, collecting error, fertilisation and poor knowledge of the biology of invertebrates are all possible reasons for the difference in the abundance of herbivores and detritivores and the total biomass of invertebrates among studies. Still, the difference in percentage distribution among the different feeding guilds in 1998, 1999 (Schneider and Grellman 2000) and 2000 in the actual study was small.
EEH
Increasing the amount of litter (i.e. increasing productivity) was expected to cause higher equilibrium density of carnivores, unchanged detritivore density, lower equilibrium density of herbivores and diminishing grazing pressure on plants; all according to EEH (Oksanen et al.
1981, Oksanen and Oksanen 2000) and apparent competition (Polis and Strong 1996). The result did not show any significant variation in invertebrate biomass, but there is a slight
20
tendency that equilibrium density of carnivores increases and herbivore density decreases with an increasing amount of litter (Figure 4 and 5). Biomass of detritivores changes when adding 100 % more litter to the system but was quite unchanged during other manipulations (Figure 4). If invertebrate carnivores are regarded as top predators and detritivores as their main resource, this tendency could be explained by the EEH hypothesis (Oksanen et al.1981, Oksanen and Oksanen 2000) and apparent competition (Oksanen et al. 1997, Schneider and Grellman 2000) between herbivores and detritivores. When litter accumulates in the system, carnivores might prefer to feed in the litter layer due to the increasing main resource
(detritivores). In the litter layer the moss eaters (Byrrhidae and Boreidae) feed, and they are therefore included in the diet of carnivores and suffers from apparent competition (Figure 5).
Results from the grazing pressure study contradicts the result from herbivore biomass study, when litter accumulates in the system, grazing activity increases, except for plots treated with 400 % litter (Figure 7, 8, 9,10 and Tabel 2). Plant tissues have lower nitrogen content than invertebrates and some plant species also produce secondary compounds (Kytö et al. 1996, Speight et al. 1999). Nitrogen and secondary compounds can fluctuate in plant tissue (Kytö et al. 1996, Stark and Grellman 2001, Larsson et al. 2000) and herbivores have to compensate for those problems (Lavoie and Oberhauser 2004). One way to compensate for lower food quality is to graze more (Lavoie and Oberhauser 2004). If plant quality became lower due to litter manipulation, increasing grazing activity could be explained by compensatory feeding by herbivores (Lavoie and Oberhauser 2004), while biomass of herbivores decreases or remains at the same level. When 400 % litter is added to the system, herbivores start to suffer from slow growth- and high mortality (Price et al. 1998, Häggström and Larsson 1995), when herbivores compensate for lower plant quality and graze more they are exposed and killed more frequently by natural enemies and decrease in density due to this.
Litter manipulation had no fertilisation effect or effects on plant community composition (Table 3), but a possible change in plant quality was not investigated.
IGP and omnivory
In the carnivore population, omnivory (Pimm 1982) and intra guild predation (IGP) (Holt and Polis 1997, Polis et al. 1989), probably occur. Some carnivore species belonging to
Carabidae, Staphylinidae and Opiliones are feeding on both dead and living invertebrates from all trophic levels (including their own level), dead plant material, and some carabids also feed on seeds (Chinery 1993). Carnivores can reduce parasitoid and herbivore populations by feeding on herbivores with parasitoids inside (Polis 1994). This type of factors, as noted in the introduction, blurs distinct trophic levels and makes it difficult to interpret carnivore impact on their resource and on their own population (Polis and Strong 1995). It also makes the likely food web of this community (Figure 1) too simplistic to explain structure and dynamics in this community. The ability of invertebrates, in this study called carnivores, to utilise resources from different types of trophic levels (omnivory) could explain why they are numerous in this community.
Ground characters
Adding litter to the ground, changes some of its characters, like moisture and sun exposure (Ljungberg 2002). These ground characters are important for carabid beetles and introduced
21
changes may effect their populations negatively (Ljungberg 2002). In a study made by Fraser and Grime (1998), the lady bird species Coccinella septempunctata was more active in warmer habitats, indicating that microhabitat is important and change in microhabitat can have negative effects on invertebrate performance. Changing ground characters and microhabitat by adding litter in this system might affect invertebrate and soil microbes performance negatively, and this type of effects was never investigated in this study.
Birds
If invertebrate carnivores are regarded as top predators, as in this study, birds are expected to be unimportant for the food web dynamics (Oksanen et al. 1981, Oksanen and Oksanen 2000). According to Schneider and Grellman (2000) birds did affect carnivore population density, but no cascading effects were detected (i.e. birds are dynamically unimportant). Birds feed on all types of invertebrates with preference for larger ones (Gunnarsson and Hake 1999), and this makes birds important as omnivores and they might affect the size structure of the invertebrate community invertebrates and the size structure becomes important if IGP occurs. The attempt to track bird predation in the actual study failed due to heavy rainfall in early July (Figure 11). The aim was to investigate if birds seemed too prefer any specific plots when foraging.
Herbivores and detritivores
There is a positive correlation between herbivores and detritivores (Figure 6), which indicates that there are one or several cooperating factors that affect their population density. This could be due to “top – down” forces by invertebrate carnivores, parasitoids and/or bird predation (Power 1992, Hunter and Price 1992, Strong 1992, Menge 1992, Persson 1999). But it could also be other factors as for example climate factors or a combination of different factors (Speight et al. 1999).
Weather
Compared to the summers of 1998 and 1999 (Schneider and Grellman 2000), the actual summer was relatively warm with low precipitation during the first part, quite cold and wet during July and ending with a cold and dry period (Figure 11). What consequence weather had for the actual invertebrate community was not investigated in this study, but temperature is important for invertebrate growth, development and activity (Chernov 1985, Speight et al.
1999). Wind can have consequences for insect movement and communication (Speight et al.
1999). Patterns of rainfall can influence long term dynamics of some insect populations (Speight et al. 1999). Rainfall affects humidity, which in combination with temperature and wind sets the conditions for microclimate, which is important for invertebrates (Chernov 1985, Speight et al. 1999). Heavy rainfall or melting snow might be able to move litter and this might have consequences for the result of this study.
22
Conclusions
The first conclusion drawn was that invertebrates classified as carnivores dominate the tundra and other invertebrates are quite rare but the reason for this pattern is unclear. It could be due to energy shunt from detritus but many other factors are probably important.
The other conclusion drawn is that the food web presented in the introduction (Figure 1) is too simplistic and the food web theory, EEH, is not adequate to describe the invertebrate food web dynamics at this tundra heath. There are many complicating factors to take into consideration when constructing a food web for the invertebrate community in this area;
omnivory, IGP, the importance of invertebrates differing in size, birds, litter composition (content of fungi, bacteria, algae etc), microclimate and life history of the invertebrate species.
References
Aunapuu, M., Dahlgren, J., Oksanen, T., Grellman, D., Oksanen, L., Olofsson, J., Rammul, U., Schneider, M., Johansen, B., Hygen, H.O. 2008. Spatial patterns and dynamic response of arctic food webs corroborate the exploitation ecosystem
hypothesis (EEH). The American naturalist 171(2): 249-262.
Breymeyer, A. 1967. Correlations between dry weight of spiders and their length and fresh weight. Bull. Acad. Pol. Sci.[Biol.] 15: 263-265.
Chase, J. M. 1998. Central-place forager effects on food web dynamics and spatial pattern in northern California meadows. Ecology, 79(4): 1236-1245.
Chernov, Y.I. 1985. The living tundra: Cambridge University Press, Cambridge.
Chinery, M. 1993. Insekter i Europa. Bonnier Alba AB, Stockholm
Diehl, S., 1993. Relative consumer size and the strengths of direct and indirect interactions in omnivorus feeding relationships. Oikos 68: 151-157
Douwes, P., Hall, R., Hansson, C., and Sandhall, Å. 1997. Insekter en fälthandbok.
Interpublishing, Stockholm.
Fraser, L.H. and Grime, J.P. 1998. Top- down control and its effect on the biomass and composition of three grasses at high and low soil fertility in outdoor microcosms.
Oecologia 113: 239-246.
Fretwell, S.D. 1987. Food chain dynamics: the central theory of ecology? Oikos 50: 291- 301.
Grellman, D. 2001. Plant responses to fertilisation and exclusions of grazers on an arctic tundra heath. In: op-Down and Bottom-up effects in a fennoscandian tundra plant community. Grellman, D. Disseration, Umeå University, Paper 2.
Gunnarsson, B. and Hake, M. 1999. Bird predation affects canopy living arthropods in city parks. Can. J. Zool. 77:1419-1428.
Hairston, N.G., Smith, F.E. and Slobodkin, L.B. 1960. Community structure, population control and competition. Am. Nat. 94: 421-425.
Hedström, L., 1995. Bestämningstabeller till svenska insektsfamiljer. Unpubl.
Henschel, J., Mahsberg, D. and Stumpf, H. 1996. Mass-length relationship of spiders and harvestmen (Aranea and Opiliones). Rew. Suisse Zool. hors série: 265-268.
Häggström, H. and Larsson, S. 1995. Slow larval growth on a suboptimal willow results I high predation mortality in the leaf beetle Galerucella lineola. Oecologica 104: 308- 315.
Holt, R. D. and Polis, G. A. 1997. A theoretical framework for intraguild predation. Am.
Nat. 149(4): 745-764.
23
Hunter, M. D. and Price, P. W. 1992. Playing chutes and ladders: heterogeneity and the relative roles of bottom-up and top-down forces in natural communities. Ecology 73:
724-732.
Jonasson, S., Michelsen, A., Schmith, I. K. and Nielsen, E.V. 1999. Response in microbes and plants to changed temperature, nutrient and light regimes in the arctic.
Ecology 80: 1828-1843.
Kytö, M., Niemela, P., Larsson, S., 1996. Insects on trees: Population and individual response to fertilisation. Oikos 75: 148-159.
Larsson, S., Ekbom, B. and Björkman, C. 2000. Influence of plant quality on pine sawfly population dynamics. Oikos 89:440-450.
Lavoie, B., Oberhauser, K. 2004. Compensatory feeding in Danaus plexippus
(Lepidoptera: Nymphalidae) in response to variation in host Plant Quality. Environ.
Entomol. 33 (4):1062-1069
Ljungberg, H. 2002. Important habitats for red-listed ground beetles in Sweden.
Entomologisk Tidskrift 123(4): 167-185.
Marquis, R. J., Welan, C. J. 1994. Insectivorous birds increase growth of white oak through consumption of leaf-chewing insects. Ecology. 75(7): 2007-2014.
MacLean, S.F. 1981. Introduction: invertebrates. Tundra Ecosystems: a comparative analysis. Cambridge university press. 509-516.
Meinander, M., Panelius, S. 1969. Bestämningstabell över Finlands dipterafamiljer.
Menge, B A. 1992. Community regulation: under what conditions are bottom-up factors important on rocky shores? Ecology. 73(3) 755-765.
Moen, J. and Oksanen, L. 1991. Ecosystem trends. Nature 353: 510.
Moran, M. D., Rooney, T. P., Hurd, L.E. 1996. Top-down cascade from a bitrophic predator in an old-field community. Ecology.77(7): 2219-2227.
Oksanen, L. and Oksanen, T. 2000. The logic and realism of the hypothesis of exploitation ecosystems. Am. Nat. 155 (6): 703-723.
Oksanen, L., Fretwell, S. D., Arruda, J. and Nemala, P. 1981. Exploitation ecosystems in gradients of primary produktivity. Am. Nat. 118. 240-261.
Oksanen, L., Aunapuu, M., Oksanen, T., Schneider, M., Ekerholm, P., Lundberg, P.A., Armylik, T., Aruoja, V. and Bondestad, L. 1997.Outlines of food webs in a low arctic tundra landscape in relation to tree theories on trophic dynamics. In: Multitrophic interactions in terrestrial systems. Gange, A. and Brown, V. Blackwell Sientific
Publications, Oxford, 351-373
Persson, L. 1999. Trophic cascades: abiding heterogenity and the trophic level concept at the end of the road. Oikos 85:385-397.
Pimm, S.L., Lawton, J.H. and Cohen, J.E. 1991. Food web patterns and their consequences. Nature 1350: 669-679.
Polis, G.A. 1991. Complex trophic interaction in desserts: an empirical critic of food web theory. Am. Nat. 138(1): 123-135.
Polis, G.A. 1994. Food webs, trophic cascades and community structure. Australian Journal of Ecology. 19: 121-136.
Polis, G.A. 1999. Why are parts of the world green? Multiple factors control productivity and the distribution of biomass. Oikos 86: 3-15
Polis, G. A., Hurd, S. D., Jackson, T., Sanches –Piñero, F. 1998. Multifactor population limitation: variable spatial and temporal control of spiders on gulf of carlifornia islands.
Ecology. 79(2): 490-502.
Polis, G.A., Hurd, S.D.1996. Allochthonous input across habitats, subsidised consumers, and apparent trophic cascades: examples from the ocean-land interface. In Food webs:
24
integration of patterns and dynamics. Edited by: Polis G. A. and Winemiller K.
Chapman hall, New York. 275-285
Polis, G.A. and Strong, D.R. 1996. Food web complexity and community dynamics. Am.
Nat. 147(5): 813-846.
Polis,G. A. and Holt, R. D. 1992. Intraguild predation: The dynamics of complex trophic interactions. Trends Ecol. Evol. 7: 151-154.
Polis, G. A., Myers, C. A. and Holt, R. D. 1989. The ecology and evolution of intraguild predation: Potential competitors that eat each other. Annual Rev. Ecol. Syst. 20: 297- 330.
Power, M. E. 1992. Top-down and bottomup forces in food webs: Do plants have primacy? Ecology. 73 (3): 733-746.
Price , P. W., Bouton, P., Gross, B. A., McPheron, J. N., Thomson, J. N. and Weis, A.
E. 1980. Interactions among three trophic levels: influence of plants on interactions between insect herbivores and natural enemies. Rewiew of Ecology and Systematic.
11:41-45.
Rammul, Ü., Oksanen, T., Oksanen, L., Lehtelä, J., Virtanen, R., Olofsson, J., Strengbom, J., Rammul, I. and Eriksson, L. 2007. Vole – vegitation interactions in an experimental, enemy free taiga floor system. Oikos. 116:1501-1513.
Rogers, L.E., Hinds, W.T. and Buschbom, R.L.1976. A general weight vs. length relationship for insects. Ann. Entomol. Soc. Am. 69: 387-389.
Rogers, L.E., Buschbom, R. L. and Watson, C. R. 1977. Length-weight relationships of shrub-steppe invertebrates. Ann. Entomol. Soc. Am. 70. 51-53.
Schneider, M. and Grellman, D. 2000. Multitrophic interactions in an invertebrate tundra food web. In: Implications of space for trophic interactions in North Fennoscandian terrestrial systems. Schneider, M. F. Dissertation, Umeå Universiy, 176-204.
Speight, M.R., Hunter, M.D. and Watt, A.D. 1999. Insects in ecosystems. In: Ecology of insects concepts and aplications. Blackwell Sientific Publications, Oxford. 169-179.
Stark, S. and Grellman. D.2002. Grellman, D. Disseration, Umeå University, Paper 4.
Strong, D. R. 1992. Are trophic cascades all wet? Differentiation and donor-control in speciose ecosystems. Ecology 73: 747-754.
Turchin, P., Oksanen L., Oksanen T., Ekerholm, P. and Hettonen, H. 2000. Are lemmings prey or predators? Nature 405: 562-565.
White, T. C. R. 1978. The importance of relative shortage of food in animal ecology.
Oecologia. 33: 71-86.
Acknowledgements
Thank you; Michael Schneider, Anders Nilsson, Doris Grellman and your friends, the Romsdal family, Maano and your field assistant.
Dept. of Ecology and Environmental Science (EMG) S-901 87 Umeå, Sweden
Telephone +46 90 786 50 00 Text telephone +46 90 786 59 00 www.umu.se
