Size-selective exclusion of mammals and invertebrates differently affects grassland plant communities depending on vegetation type
Wang, Xiaowei; Schutz, Martin; Risch, Anita (2021), Size-selective exclusion of mammals and invertebrates differently affects grassland plant communities depending on vegetation type, Dryad, Dataset, https://doi.org/10.5061/dryad.pg4f4qrnc
Human-caused loss of vertebrate and invertebrate animals, defaunation, is increasing, and potentially affects plant community structure of diverse grassland ecosystems worldwide. We experimentally simulated defaunation using size-selective fences to progressively exclude large-, medium- and small-sized mammals, and invertebrates from two subalpine vegetation types in the Swiss National Park (SNP): intensively grazed short-grass and moderately grazed tall-grass vegetation. We assessed plant community properties yearly from 2009 to 2013, and examined treatment effects on plant community structure in the two grassland types. In the short-grass vegetation, the exclusion of large mammals increased total plant biomass, while the exclusion of large and medium-sized mammals increased total, grass and forb biomass compared to when all animals had access. These increases became stronger when also invertebrates were excluded. The exclusion of all mammals and invertebrates increased biomass of grasses by 205%, forbs by 100% and total plant biomass by 118% compared to when all animals had access, hence enhancing relative biomass of grasses from 43.6% to 60%, changing plant species composition and lowering richness of forbs by 16%, the number of plant families by 13% and family-level Shannon diversity by 23%. In contrast to these significant community-level responses found in the short-grass vegetation, there was no evidence that the size-selective exclusion of animals altered the plant community structure of the tall-grass vegetation. The contrasting results were due to the difference in plant community composition prior to our experiment, which were related to differences in quantity and quality of forage and in grazing intensities of herbivores between the two grassland types. Synthesis. Our results showed that different-sized animals, in particular large mammals and invertebrates, contributed to maintain the plant community structure in the short-grass vegetation, highlighting the importance of multiple, functionally different animal groups for ecosystem functioning and stability. In contrast to the short-grass vegetation, we could not detect such a top-down control by animals in the tall-grass vegetation. Our results suggest that potential defaunation effects on grassland plant community structure depend on the degree of grazing pressure release and grassland vegetation type.
1 Study area
The study was conducted in subalpine grasslands in the Swiss National Park (SNP). The SNP is located in the central Alps, and covers an area of 172 km2 ranging from 1350 to 3170 m above sea level (a.s.l.), with mean annual precipitation of 826 ± 112 mm and mean annual temperature of 0.9 ± 0.5°C (mean ± SD, 2009 - 2013; MeteoSchweiz, 2014). Forests (50 km2), alpine grasslands (33 km2) and subalpine grasslands (approximately 3 km2) cover around 50% of the SNP (Risch et al., 2013), the rest is dominated by rock and scree. The underlying bedrock of the subalpine grasslands is dolomite.
Since its foundation in 1914, the SNP has been protected from human disturbances (no livestock grazing, no hunting, no fishing, no camping, or no off-trail hiking). The subalpine grasslands are comprised of homogeneous patches (usually >1 ha) of short- and tall-grass vegetation, which developed mainly because of differences in land-use history and grazing by domestic livestock and wild ungulates over several centuries (Schütz et al., 2003, 2006). Soon after banning cattle in 1914, red deer (Cervus elaphus L.) re-migrated into the SNP and started to preferentially graze where cattle formerly rested overnight and soils became enriched in nutrients, creating what we today call short-grass vegetation (Schütz et al., 2003). Tall-grass vegetation developed where cattle frequently grazed, but not rested before the foundation of the SNP. These areas became depleted of soil nutrients. The vegetation, therefore, is less nutrient-rich, hence, red deer graze these areas with much lower intensity (Schütz et al., 2003).
The aboveground animals found in our study area can be divided into four groups (Risch et al., 2013): (1) large mammalian herbivores (30–150 kg; mainly red deer and chamois Rupicapra rupicapra), (2) medium mammals (3–6 kg; e.g. alpine marmot Marmota marmota and mountain hare Lepus timidus), (3) small mammals (30–100 g; small rodents such as Clethrionomys spp., Microtus spp. and Apodemus spp.), and (4) invertebrates (< 5 g; e.g. grasshoppers, caterpillars and leafhoppers). At present a total of 26 species of large to small wild mammals can be found in the SNP (see Risch et al., 2018), with large mammalian predators (wolf, bear or lynx) being absent or non-resident during our experiment. Reptiles, amphibians and birds are scarce in the subalpine grasslands. See Risch et al. (2018) for a full list of invertebrate species captured in the SNP in summer 2013.
2 Experimental design
In early May 2009, we established eighteen fencing setups (referred to as random factor 'Fence' in the Statistical Analysis section) in six subalpine grasslands (random factor 'Grassland') located across the SNP at elevations ranging from 1975 to 2300 m a.s.l. There were two fencing setups per vegetation type (short-grass vs. tall-grass) in each of the three large grasslands, and one fencing setup per vegetation type in each of the three small grasslands. Hence, we established nine fencing setups in short-grass vegetation and nine in tall-grass vegetation in a paired design (Fig. S1). The fencing setups were left in the field for five consecutive growing seasons (May 2009 - October 2013), but to protect them from snow damage and avalanches they were temporarily dismantled in late October every year, and reconstructed in early May of the following year, immediately after snowmelt.
For a detailed fencing protocol see Risch et al. (2013). Briefly, each fencing setup consisted of a 2.1 m tall and 7 × 9 m large main fence and we established five 2 × 3 m plots that simulated the size-selective exclusion treatments (Fig. S1). The 'LMSI' plot was located at least five meters away from the main fence (i.e., control plot; 2 × 3 m) and gave access to large (L), medium (M) and small (S) mammals, and invertebrates (I). The main fence consisted of wooden posts and electrical wires mounted at regular intervals between 0.5 and 2.1 m to keep large mammals out. Note that the bottom wire (0.5 m) was not electrified to allow smaller mammals to enter safely. Within each main fence, we randomly established four 2 × 3 m treatment plots one meter from the main fence line to avoid edge effects and separated by one meter walkways: (1) the ‘MSI’ plot remained unfenced and gave access to medium (M) and small (S) mammals and all aboveground dwelling invertebrates (I); (2) the 'SI' plot was surrounded by an electrical fence with 10 × 10 cm mesh size and allowed small (S) mammals and invertebrates (I) to enter; (3) the 'I' plot was surrounded by 2 × 2 cm mesh-sized metal fence, double-folded at the bottom 50 cm, which only allowed invertebrates (I) to enter; and (4) the 'None' plot was surrounded by a 1.5 × 2 mm mesh-sized mosquito net and covered with a mosquito mesh-lined roof to exclude all vertebrate and invertebrate animals (Fig. S1). We built additional six 'micro-climate control' exclosures (one in each of the six main grasslands; see Risch et al., 2013, 2018 for the construction details) with a comparable micro-climate to the 'None' plots, but also a comparable feeding pressure by invertebrates to the 'I' plots. As discussed in detail in Risch et al. (2013, 2018), differences in plant and soil properties between the 'I' and the 'None' treatments were not due to the exclosure design (mesh and roof) of the 'None' exclosure, but a function of animal exclusions.
3 Vegetation characteristics
We collected plant data in a total of 90 plots (18 × 5). We quantified vegetation characteristics yearly at peak biomass (July) in predetermined, randomly assigned 1 × 1 m subplots within each 2 × 3 m plot over five consecutive growing seasons (2009 - 2013). We assessed plant community composition by identifying all vascular plants to species level. We estimated percentage cover of each species and assigned each species to one of five plant functional groups: forbs, grasses, sedges, legumes, or woody species. We measured aboveground biomass (g/m2) of each plant functional group (see Table 1) using the canopy intercept method (Frank & McNaughton, 1990). Total aboveground plant biomass (i.e., biomass of live shoots, 'BM.shoots') was defined as the sum of biomass of all five plant functional groups. As woody species were rare and their biomass negligible, we did not further analyse their responses to the exclusion treatments.
For each plot and study year, we calculated different plant diversity measures: overall species richness (total number of plant species), Shannon diversity (exponential of Shannon entropy), plant family richness (number of taxonomic families), family-level Shannon diversity, and richness of each plant functional group. We also calculated Rao’s quadratic entropy (Laliberté, Legendre, & Shipley 2014), which accounts for both species relative abundance and traits, as an overall measure of plant functional diversity (FDq). Species traits that we considered here were taxonomic family, functional group, and ecological indicator values for moisture, pH, nutrients, light and temperature (Landolt et al., 2010; Table S4). Life span was not included in the trait dataset because over 95% of species recorded in our study are perennials.
The dataset was used for analyses in the following journal article: Wang, X., Schütz, M., & Risch, A. C. (2021) Size-selective exclusion of mammals and invertebrates differently affects grassland plant communities depending on vegetation type, Journal of Ecology.
Users of the dataset should cite the journal article AND the Dryad Dataset Repository. To correctly understand and use the dataset, it is suggested to read the METHODS of this research (see above).
For further information, please contact: email@example.com
Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung, Award: 31003A_122009/1, 31003A_140939/1
National Natural Science Foundation of China, Award: 41401231, 41630755