SEARCH

SEARCH BY CITATION

Keywords:

  • alpine meadow;
  • bottom-up and top-down control;
  • global warming;
  • Tibetan Plateau;
  • trophic interaction

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

1. It is well known that climate change alters abiotic factors (temperature and water availability) that directly affect ecosystem properties. However, less is known about the indirect impacts of climate change on ecosystem structure and function. Here, we show that experimental warming may deteriorate ecosystems via trophic interactions.

2. In a Tibetan alpine meadow, plant species composition, size, coverage and above-ground biomass were investigated to reveal the effect of artificial warming (c. 1 °C mean annual temperature at the soil surface), which was accomplished using warmed and ambient open top chambers. In addition, rodent damage to plants was assessed.

3. The dicot forb silverweed Potentilla anserina increased significantly, while other species groups remained unchanged or decreased in plant community dominance rank after 2 years of artificial warming. The change in community structure was attributed to the difference in biomass allocation and growth form among species.

4. In the third year, plateau zokors Myospalax fontanierii, a widespread rodent herbivore, damaged plants in the warmed chambers, while leaving plants in the ambient chambers mostly undamaged. Above-ground biomass was found to be smaller in the warmed chambers than the controls in the third year, in contrast to the trend of the first 2 years. In addition, zokor burrow density was positively correlated with silverweed biomass and its dominance within communities, which was consistent with findings of independent field investigations that silverweed-dominated plots were more likely to be visited and damaged by the zokors than sites-dominated by grass species.

5.Synthesis and applications. The top-down negative effect of zokor damage on above-ground biomass in the warmed chambers was induced by the bottom-up effect of changes in species composition and community structure on zokor foraging behaviour, which were driven by artificial warming. Such trophic interactions may invalidate some predictions of ecological effects by current species-climate envelope models. Furthermore, because management measures including increasing the water table, planting grass and moderate cattle grazing may prevent silverweed dominance, we suggest that these interventions could be employed to control zokor damage in alpine meadows that are predicted to be drier and warmer in the future.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

The impact of global warming is likely to increase during the next several decades (Solomon et al. 2007). Biological responses to recent temperature increases are already evident and include significant shifts in phenology and species geographic distribution (Chapin et al. 1995; Root et al. 2003). Global warming has led to ecosystem degradation in terrestrial, coastal and aquatic biomes (Zhang, Douglas & Leatherman 2004; Lawrence et al. 2007; Reynolds et al. 2007). Dry lands have become more fragile with increasing losses of water and species (Schlesinger et al. 1990; Reynolds et al. 2007). These degradations in ecosystem function are primarily driven by abiotic factors directly related to temperature. However, global warming may also lead to changes in ecosystem function indirectly via trophic interactions (sensuSchmitz et al. 2003; Suttle, Thomsen & Power 2007).

Plant productivity has been shown to increase with temperature in high latitude and altitude areas such as northern boreal forest, arctic, Antarctic and alpine tundra (e.g. Field et al. 1998; Peñuelas & Filella 2001). In particular, primary productivity and nutrient cycling rates of alpine meadows and arctic tundra are expected to increase with increasing temperatures (Arft et al. 1999; Dormann & Woodin 2002). Because these areas are characterized by low temperature that can limit plant growth and soil nutrient mineralization, global warming will alleviate these limitations and increase ecosystem function and service (e.g. cattle production in alpine pastures) of these biomes.

One of the unknown consequences of increasing plant primary productivity is the effect on trophic interactions with other community members. On the one hand, increased temperature in cold regions often increases plant growth, which may decrease leaf quality (e.g. Cornelissen et al. 2007) for small leaf herbivores like aphids, but may also increase resource abundance for other herbivores. For example, large herbivores are usually not sensitive to minor changes in leaf quality of grass species; therefore, their abundance could increase with increasing primary productivity of rangelands in response to warming via a bottom-up effect. Similarly, rodent species usually rely on plant seeds and roots, and if production of these plant tissues is enhanced by increased temperature, then we would expect rodent populations to benefit from increasing plant productivity. On the other hand, once the bottom-up effect is realized, the abundance of consumers (including herbivores and predators) may be enhanced and their feeding, mating, and reproductive behaviours may be changed (Barton & Schmitz 2009), thereby exerting top-down control. Recently, experimental warming has revealed varied indirect effects of predators on plants and changed trophic relationships between predators in old-field grassland food webs (Barton, Beckerman & Schmitz 2009; Barton & Schmitz 2009). These experiments primarily focused on the effects of warming on small arthropods such as spiders and grasshoppers; the effect of vertebrates has seldom been addressed under experimental conditions.

The Tibetan Plateau is an extensive alpine zone, part of which is highly productive, with meadows forming the fifth largest husbandry base in China (Xiang et al. 2009). Because this plateau is characterized by cold weather all year, it is predicted to be very sensitive to global warming (McCarthy et al. 2001). Specifically, this area will be likely to experience a greater than average increase in temperature in the near future (Solomon et al. 2007). As noted above, in such cold and moist areas, enhanced temperature may facilitate plant growth and primary production, although the effect depends on cattle grazing (Klein, Harte & Zhao 2004, 2007). In addition to cattle, the plateau zokor grazers Myospalax fontanierii Milne-Edwarda are also abundant in these grasslands. The plateau zokor is a specialized subterranean rodent herbivore that is distributed across the Tibetan Plateau in farm and alpine meadows (Zhang, Zhang & Liu 2003). Population density of this rodent has increased rapidly in recent years due to overgrazing by cattle or climate change (Xiang et al. 2009). The traditional viewpoint is that there is a positive feedback between rodent density and vegetation: reduction of plant coverage, density and height and soil moisture facilitates the rodent invasion that in turn decreases vegetation cover and plant productivity (Zhang, Zhang & Liu 2003; Xiang et al. 2009). If a warmer climate leads to increasing plant growth and density, that would help the plant community resist rodent invasion. Therefore, we hypothesize that increased temperature in a cold high elevation meadow setting will lead to increasing plant productivity, allowing the plant community to resist damage by the plateau zokor.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

Study site

This study was conducted in an alpine meadow in Hongyuan County, Sichuan Province, on the eastern Qinghai-Tibetan Plateau (31°50′–33°22′N, 101°51′–103°23′E). The altitude is c. 3500 m. The annual mean temperature is c. 0·9 °C with the maximum and minimum monthly means ranging from 10·9 to −10·3 °C in July and January, respectively. The annual mean precipitation is between 650 and 730 mm, fluctuating greatly among years, 80% of which occurs during May to August. The annual mean duration of snow cover is 76 days.

The meadow has over 80% vegetation cover, and average maximum plant height is about 30 cm. It is mostly dominated by sedges such as Sichuan kobresia Kobresia setchwanensis Hand.-Mazz. and alpine korbresia Kobresia pygmaea Clarke, and grasses like tufted hairgrass Deschampsia caespitosa L., bunchgrass Agrostis clavata Trin, and wild rye Elymus nutans Griseb. Forb species, including anemone Anemone trullifolia var. linearis Hand.-Mazz., silverweed Potentilla anserina L., and saw-wort Saussurea nigrescens Maxim., are also abundant in the meadow. The most dominant grass and forbs species are the hairgrass and anemone, respectively. Silverweed is widely distributed across the meadow as a subordinate species with a clonal growth form and large rootstock (known as droma), which is consumed as food by the local Tibetan people. Silverweed is one of the favourite foods of the plateau zokor (Zhang 2000).

The soil is characterized by high organic content (215–280 g kg−1) and low total N (6–10 g kg−1) and available P (3–7 mg kg−1). Over the past decades, alpine meadows in the Qinghai-Tibetan Plateau have experienced severe soil degradation due to an increased rodent population, intensive farming activities and climate change (Li et al. 2009). The pasture has been under intensive grazing for decades.

The pasture has also been used by plateau zokors, rodents morphologically and behaviourally similar to pocket gophers (Cratogeomys; Andersen 1987). Plateau zokors spend their lives solely in underground burrow systems, and excavate tunnels from deep nests to the subsurface for foraging (Zhang, Zhang & Liu 2003) by using limbs and heads to dig and push the loosened soil to the surface; they also use strong incisors to cut roots and drag whole plants into the deep tunnel system (Zhang, Zhang & Liu 2003) and make mounds by digging, wriggling, pushing up and mixing soil while foraging the plants. It was estimated that they occupied a range of c. 3·8 × 106 ha at an average density of 15 per ha in some damaged areas (Zhang 2000; Zhang, Zhang & Liu 2003).

The warming experiment

In the year before starting our experiment, we fenced a 100 × 50 m plot in a meadow with a fairly uniform mixture of vascular plant species such that more than 90% of the plant species (harvested in this study) could be found in any 25 × 25 cm grid square. No rodent damage was found within the fenced area. Cattle grazing occurred before fencing, but was precluded thereafter.

In June, 2007, twenty 2 × 2×2 m (height) open top chambers (OTCs) were randomly deployed (with a minimum distance of 3 m between them) within the fenced area. The chamber sides were covered with thin (<0·1 mm) steel screen with a mesh size of 0·2 × 0·2 mm. Half of the chambers were also covered with poly-carbonate screen with a transparency of over 90%. We refer to these two types of OTCs as the ambient, unwarmed, control chambers and the warmed chambers, respectively. Each OTC was sunk 10 cm into the soil and firmly stabilized with vertical steel poles that were held up by poured concrete to withstand the windy weather at the study site. Each OTC covered 4 m2, generally larger than those in the other studies conducted in grasslands (Klein, Harte & Zhao 2004, 2007; Jägerbrand et al. 2009) with the aim of incorporating the activities of some invertebrates such as insects and spiders.

Thermometers (DS1921G; Maxim integrated products, Sunnyvale, California, USA) were placed at the plot centres and showed that mean annual temperature at the soil surface and at 30 cm above the surface was enhanced by the poly-carbonate screen by 1·3–1·6 °C (see Table S1, Supporting information). The temperature increase was smaller at 5 cm below the soil surface than at the soil surface, and was smaller during the growing season (April to September) than the non-growing season. Soil moisture was also recorded bimonthly during the experiment except for the freezing period (from November to March). It was slightly higher (but not systematically and significantly so) in the warmed chambers than in the controls (see Fig. S1, Supporting information). In contrast, a short-term record showed that relative humidity at 30 cm above the ground was generally lower in warmed OTCs than in the controls, although there was large variation with time (see Fig. S2, Supporting information).

In each OTC, one 1·5 × 1·5 m subplot was set out for plant and soil measurements and sampling, but the central 1 × 1 m part (surrounded with a red steel frame) was set only for plant community survey and biomass harvest. At the end of the growing seasons in 2007, 2008 and 2009, plant species composition, plant height and coverage, and above-ground biomass (AGB) were investigated before leaves turned yellow in September. We recorded average height, density and cover for each plant species for the central 1 m2 part of each plot and chamber. A 1-m2 quadrat frame consisting of 10 × 10 cm grid squares was used to facilitate the cover and density estimates; the plant growth parameters were recorded for each grid and then pooled for each OTC. After the plant community survey, we clipped above-ground plant parts, which were dried at 65 °C for 48 h and then weighed. The biomass was grouped into four species groups: grass, sedge, silverweed and other forbs.

While we were investigating the plant community in September 2009, rodent damage was found in the OTCs. The damage was recognized to be primarily due to plateau zokors that were observed, trapped and identified, but we could not exclude the involvement of other rodent species. In order to estimate the magnitude of damage, we recorded the number of burrows the zokors dug and the area of mounds piled by the rodents as they dug tunnels into OTCs. Mound size was calculated assuming that their shape is approximately an ellipse, using the formula παβ/4, where α and β were the long and short axes of the ellipse, respectively. However, we did not examine underground tunnels and we did not estimate the underground damage. Only the number of burrow entrances and mounds on the surface of the soil were recorded and measured.

The relationship between plant traits and rodent damage

To further investigate the relationship between damage caused by plateau zokors and plant community structure, a field investigation and a field experiment were independently conducted in 2009 outside the fenced site. The field investigation was carried out in two large plant communities (more than 0·5 ha each), both on moist soil sites with similar levels of grazing, but with contrasting community structures. One community was dominated primarily by grass and sedge species such as the wild rye, the bunchgrass and bulrush Scirpus pumilus Vahl., followed by the forb species such as alpine bistort Polygonum viviparum L. var. viviparum Li, and edelweiss Leontopodium spp. The total vegetation cover was about 80% and the average plant height was about 20 cm. The other community was dominated primarily by forb species like silverweed and plantain Plantago depressa Willd., followed by thyme-leaved sandwort Arenaria serpyllifolia L., and edelweiss. The plant cover was about 75% and the average plant height was about 15 cm. In each community, ten 5 × 5 m plots were randomly located to estimate the level of zokor damage at the end of September 2009.

The complementary field experiment was initiated in early May of 2009. Seedlings of P. anserina and seeds of E. nutans were planted within 60 2 × 2 m plots (30 for each) that were randomly distributed in a fenced area. For Potentilla plots, 100 seedlings with heights of no more than 5 cm were evenly distributed in each plot. For Elymus plots, 10 g of seeds were sown uniformly into the soil per plot at the same time as the Potentilla plots were planted, using a spacing of 20 cm between rows. At the end of September 2009 when the plants were above 20 cm in height, rodent damage was assessed. We recorded burrow number and area using the same methods described above for the OTCs.

Data analysis

We used importance values and relative AGB values to characterize the community status of four different species groups: grass, sedge, silverweed and other forbs. The importance values were calculated as the average of relative plant density and relative plant cover, which were further calculated as the total density (or cover) of any species group divided by the total density of all species groups. As a complementary analysis, the relative AGB value was also calculated for each species group as the biomass ratio of the specified species group to all species groups.

All data were tested for normality before analyses. Two-way anova and post hoc Tukey tests were conducted to examine the effects of experimental warming and life-form on importance values and relative above-ground plant biomass. Student’s t test for independent variables was employed to determine the difference in rodent damage (burrow number and mound area) between warmed and ambient OTCs, and also the difference in rodent damage between the two contrasting communities in both the field survey and the field experiment. In addition, linear regressions were used to determine the relationship between the importance value of silverweed and rodent damage level among the OTCs. All these data analyses were performed with statistica for Windows (StatSoft Inc. 2000).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

Warming effect on above-ground plant biomass and community structure

There were large variations in primary productivity, estimated as AGB, among years. It was generally higher in 2007 than 2008 or 2009 (Fig. 1a). In particular, AGB was significantly greater in the warmed chambers in 2007 than all the other treatments in 2008 and 2009 (< 0·01 for all comparisons, post hoc Tukey test; Fig. 1a). However, the warming effect was not significant in 2007 and 2008 (P = 0·323 and 0·260, respectively; post hoc Tukey test), although AGB was slightly higher in warmed OTCs than in the controls. In contrast, primary productivity was significantly smaller in warmed OTCs than in ambient chambers in 2009 (P = 0·01, post hoc Tukey test).

image

Figure 1.  Total above-ground plant biomass (AGB) (a) and relative AGB values for four species groups between warmed (black columns) and ambient chambers (white columns) in three study years of 2007 (b), 2008 (c) and 2009 (d). Significance levels of the differences between treatments are denoted by *< 0·05 and **< 0·01 above the columns. Error bars indicate 1 SEM. = 10 for each treatment.

Download figure to PowerPoint

The AGB responses to experimental warming were not consistent among the four species groups, as indicated by the variation in relative AGB values (Fig. 1b–d). Temperature did not have a significant effect on overall relative AGB; however, the species group effect and its interaction with the warming effect were significant in 2007 and 2008 (Table 1). During the first 2 years, silverweed had significantly higher relative AGB value in the warmed OTCs than the controls (= 0·033 and 0·009 for 2007 and 2008, respectively; post hoc Tukey test), while the other species groups either showed a non-significant decrease or remained unchanged (Fig. 1b,c). In contrast, this tendency was somewhat reversed in 2009, when there was no significant difference in silverweed AGB between the warmed and the control chambers, while the previous lack of significant differences for the other species groups during 2007 and 2008 was repeated (Fig. 1d).

Table 1.   Results of two-way analyses of variance showing the effect of temperature (warmed vs. ambient), life-form (forbs, grass, sedge and silverweed) or year (2007, 2008 and 2009) and their interactions on above-ground plant biomass (AGB), relative AGB value and importance value of different species groups in three study years (2007–2009) during the warming experiment
 SourceSSd.f.MSFP
AGBYear181 706290 85333·8680·000
Temperature2791127911·0400·312
Year × Temperature36 992218 4966·8950·002
Error144 858542683  
Relative AGB value
 2007Temperature0·00010·0000·0001·000
Life-form0·49630·16575·8180·000
Temperature × life-form0·03330·0115·0170·003
Error0·157720·002  
 2008Temperature0·00010·0000·0001·000
Life-form1·26030·420114·8060·000
Temperature × life-form0·07330·0246·6950·000
Error0·263720·004  
 2009Temperature0·00010·0000·0001·000
Life-form0·53930·18084·0790·000
Temperature × life-form0·02930·0104·5290·006
Error0·154720·002  
Importance value
 2007Temperature0·00010·0000·0001·000
Life-form0·31530·105172·9280·000
Temperature × life-form0·02030·00711·2010·000
Error0·044720·001  
 2008Temperature0·00010·0000·0001·000
Life-form0·99030·330228·2130·000
Temperature × life-form0·04130·0149·4870·000
Error0·104720·001  
 2009Temperature0·00010·0000·0001·000
Life-form0·30230·10182·5900·000
Temperature × life-form0·01730·0064·5320·006
Error0·088720·001  

Variations in importance values were consistent with those of relative AGB values in different years. There was not a significant effect of warming on the importance values, whereas the species group effect and its interaction with warming effect were significant in 2007 and 2008 (Table 1; Fig. 2). Experimental warming significantly increased the importance value of silverweed in the first 2 years (< 0·001 and P = 0·008, post hoc Tukey test), but the values were decreased or remained unaffected in the other species groups (Fig. 2a,b). In contrast, no significant effect was observed in any of the four species groups in 2009, although the value slightly decreased in the forbs (Fig. 2c).

image

Figure 2.  Variation in importance value among four species groups in three study years of 2007 (a), 2008 (b) and 2009 (c). Significant differences between warmed (black columns) and ambient chambers (white columns) are denoted by *< 0·05, **< 0·01 and ***< 0·001 above the columns. For calculations of the importance value, see text. Error bars indicate 1 SEM. = 10 for each treatment.

Download figure to PowerPoint

Rodent damage in relation to plant community composition

The extent of rodent damage was very different between the warmed and ambient OTCs. Nine warmed OTCs were visited and damaged by plateau zokors, as suggested by the presence of burrow entrances and mounds, while only three control chambers were damaged by the rodents. Burrow density and percentage of area covered by mounds were significantly higher in the warmed than the control OTCs (t = 4·598, < 0·001 and t = 3·420, < 0·01, respectively; Fig. 3a,b), with the burrow density about seven times higher and the percentage area covered by mounds about 10 times higher for the warmed OTCs as for the control chambers.

image

Figure 3.  Damage by plateau zokors as indicated by burrow density (a) and percentage of damage area (b) between warmed and ambient chambers (= 10 for each treatment) at the third year of the warming experiment.

Download figure to PowerPoint

Rodent damage was related to the community status of silverweed. Burrow density was found to be significantly and positively correlated with above-ground biomass and relative AGB value (all < 0·05; see Fig. 4a,b), and the importance value of silverweed (see Fig. 4c).

image

Figure 4.  Relationships between burrow density and (a) above-ground plant biomass (AGB) and (b) relative AGB of silverweed Potentilla anserine in 10 warmed chambers. Both the regressions were significant (= 10, < 0·05).

Download figure to PowerPoint

The field investigation revealed that forb-dominated communities were more likely to be visited and damaged by zokors than grass-dominated communities (Fig. 5a). Consistently, the field experiment with planted grass and silverweed communities revealed that the silverweed plantation was highly preferred by the rodents (Fig. 3d) relative to the grass plantation (Fig. 5b), such that approximately half of the plantation was destroyed by the burrows and mounds. In contrast, rodent damage was very low in the grass plantation (<5% of total area, Fig. 5b).

image

Figure 5.  Damage by plateau zokors as indicated by percentage of damage area (a) between grass-dominated and forbs-dominated communities in a field survey (= 10 for each treatment) and (b) between grass and silverweed monospecific plantations in a field experiment (= 30 for each treatment). The difference between treatments was highly different (< 0·001) in all panels. Error bars indicate 1 SEM.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

The hypothesis that warming would help alleviate rodent damage by increasing productivity was not supported by the data. In contrast to this prediction, after 3 years of artificial warming, we observed a significant adverse effect of temperature increase on above-ground biomass, principally through the mechanism of trophic interactions. We found that experimental warming changed species composition and community structure during the first 2 years such that silverweed increased in total biomass and relative abundance. The change in community structure induced invasion of the plateau zokor, which during the third year of the experiment damaged the vegetation of warmed chambers by uprooting plants, feeding on plant storage organs (tubers, rhizomes, and roots), and covering plants with soil piled up around their burrow entrances, all of which in turn significantly decreased plant productivity.

Thus, the results of our study indicate that global warming may degrade ecosystems through trophic interactions, specifically interactions between plant community composition and herbivore activities. Even when plant productivity is enhanced by global warming, predictions need to take into account trophic interactions among plants and herbivores.

Top-down vs. bottom-up effects

Top-down and bottom-up effects on the study system were both observed in this study. The difference in AGB between warmed and control OTCs decreased as the experiment progressed during the first 2 years, possibly due to acclimation of plant response to warming. The change in species diversity (evenness in this study; e.g. silverweed increased dominance while other species groups decreased or remained unchanged) after warming might also have contributed to the decrease. However, the difference in AGB between treatments was reversed in the third year, apparently due to zokor damage. Strong top-down effects on the plant community in the warmed chambers were principally due to the special traits of plateau zokors in the study region. Insect populations, with the exception of large outbreaks, usually cause limited damages to plants by removing parts of plants without killing them (e.g. Barton, Beckerman & Schmitz 2009; Barton & Schmitz 2009). Large grazers, such as yaks and horses in the alpine plateau, remove only above-ground parts of plants, allowing re-growth. In contrast, plateau zokors mostly kill the plants as a whole, eliminating the possibility of re-growth or compensatory response. Specifically, we observed that many silverweed plants were uprooted and scattered within the tunnels of plateau zokors in some OTCs and other places outside our study site. This probably led to the observed decrease in dominance of silverweed in the warmed OTCs compared to ambient controls in the third year of the study.

Another important factor that makes top-down effects possible is the relatively high plant productivity that sustains rodent density in our study area; an AGB of >300 g m−2 is comparable to many prairies and old-field grasslands (e.g. Polley, Wilsey & Derner 2007; Bakker et al. 2009), and greater than many dry or short steppe grasslands and deserts (e.g. Ni et al. 2007; Muldavin et al. 2008). The top-down effects of herbivores on plant production and composition of ecosystems are often stronger in highly productive areas (Chase et al. 2000), where rodents have been shown to exert significant top-down control over vegetation cover and diversity of prairie, grassland, tundra and forest plant communities (Moen et al. 1993; Meserve et al. 2003; Goheen et al. 2004). In contrast, in a low-productivity aridland such as the Chihuahua desert, the abundance of consumers is often constrained, limiting their influence on plant community structure (Báez et al. 2006). Furthermore, as noted, warming should have changed species diversity and hence the strength of the top-down control of rodents on AGB, a suggested by Duffy (2002) and Hillebrand, Bennett & Cadotte (2008).

Mechanisms responsible for bottom-up effects involve two consecutive responses to experimental warming. First, silverweed gained more dominance in the community after 2 years of warming, although the other forbs were unaffected or even decreased with experimental warming. In other warming experiments conducted in subarctic, arctic, high arctic and alpine tundra and meadows, forb species generally became more abundant while grass species declined in coverage (Chapin et al. 1995; Arft et al. 1999; Klein, Harte & Zhao 2007; Jägerbrand et al. 2009). In the present study, the total community status did not significantly change if only two species groups (dicot and monocotyledonous herbaceous) were followed.

The increase in silverweed may be attributed to temperature increase and possibly associated increases in water loss. Although we did not find a systematic and significant difference in soil moisture between the warmed and ambient OTCs, relative humidity was slightly lower in the warmed OTCs (see Fig. S2, Supporting information). This probably led to increased transpiration, which would favour the species with greater water uptake ability. The silverweed may have greater water uptake because it has a taproot much deeper than shallow roots of the dominant grass and sedge species. Although temperature increase in a cold-limited ecosystem such as our study area should improve plant growth, it may also intensify interspecific competition. Silverweed has two adaptations that are likely to allow it to outcompete other species with warmer conditions, including tubers attached to rhizomes that provide energy and nutrients, and clonal growth form that allows it to fill empty space in communities.

The second of two consecutive responses responsible for bottom-up effects is that the change in community composition induced the invasion and foraging by zokors. Plateau zokors have broad diets, consuming both roots and shoots of annual and perennial grasses, forbs, and a few shrubs (Zhang 2000), but they forage selectively. They prefer areas with a soft soil layer, high primary productivity (Wang et al. 2000; Zhang 2000) and forbs with succulent below-ground storage organs (such as Notopterygium forbesii Boiss, P. anserina, and Morina chinensis Diels; Zhang 2000). In this study site, silverweed is the only species favoured by the zokor. Silverweed had fewer but larger first-order roots than the dominant grass and sedge species (1–4 and >10 per individual, respectively; S. Sun pers. obs.). Therefore, with increasing silverweed dominance, the root concentration should become greater in the warmed chambers than the control ones. The rodents would be more likely to visit and forage within the warmed OTCs if the difference is strong enough and if they can sense and locate the roots.

One could argue that zokors visited the warmed chambers more than the ambient chambers because of warmer temperatures. However, the zokors mostly live underground and are unlikely to respond to the difference in the air temperature above the soil surface and there is no evidence showing that the zokor can detect the very slight difference (0·3–0·8 °C) in soil temperature between warmed and unwarmed plots. In addition, predation by upland buzzards Buteo hemilasius Temminck & Schlegel and saker falcons Falco cherrug Gray (Zhang & Liu 2003; Zhang, Zhang & Liu 2003) might be inhibited by the 2 m tall sides of the OTCs, restricting the ability of these birds to swoop down upon the zokors. However, these bird species are rarely observed at the study site, and the ambient OTCs had sides just as tall as the warmed OTCs. Furthermore, data from the field investigation and field experiments showed that zokors preferred plots with high abundance of silverweed (like the warmed OTCs), even though these plots were neither warmed nor surrounded with 2 m tall sides. Accordingly, there is no evidence showing that the zokors visited the warmed OTCs simply because of their physical setting.

Conclusions

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

In summary, we have demonstrated that plant community structure and above-ground biomass of the alpine meadow was significantly affected (increased and then decreased) by experimental warming through superior competitive ability of silverweed under warmed conditions followed by trophic interactions between silverweed and the zokor. We note that many silverweed plants were badly damaged by excavation of, or burial in, zokor mounds, and that plants had not recovered by summer 2010. Although this suggests long-lasting damage by the zokor, the community may gradually recover if the monocot species invade and occupy the damaged areas. Nevertheless, the long-term balance between the zokors and the silverweed is beyond the scope of this paper. Also, not all alpine meadow communities on the Tibetan Plateau are dominated by silverweed, and thus will not necessarily experience similar trophic interactions under global warming.

Our study results complicate previous predictions about ecosystem function (e.g. primary productivity) in cold, high elevation meadows with a warmer climate, because previous models have focused on the effects of rainfall and temperature on plant species distribution and growth, whereas this study supports the idea that studies of global change should move beyond vegetation mapping (Schmitz et al. 2003) and address the issue of trophic interactions. This also indicates that ecosystem degradation could be more serious than predicted when only the direct effects of climate change are taken into account, even in ecosystems such as low-temperature and high soil-moisture regions, where global warming is predicted to increase primary productivity. Trophic interactions and associated underlying mechanisms under global change need to be extensively studied so as to decrease the uncertainty of prediction about ecosystem functioning.

In addition, our results may have implications for ecosystem management to alleviate the possible adverse effects of global warming in alpine meadows. First, the trophic interaction revealed between silverweed and the zokor suggests that preventing potential dominance of silverweed may help decrease rodent damage. Recent declines in the soil water table, often attributed to ditch digging and climate change, have greatly changed vegetation cover, community composition and diversity (Xiang et al. 2009) by allowing the drought tolerant silverweed to replace the more water demanding sedge and grass species. Zokor damage is also reported to be increasing (Xiang et al. 2009). Therefore, ditch-filling and planting grass could be important management actions to restore damaged grassland and avoid more zokor damage, as suggested by other researchers (e.g. Zhang, Zhang & Liu 2003; Li et al. 2009; Xiang et al. 2009). Secondly, overgrazing by cattle can decrease the abundance of sedge and grass species, thus exacerbating the increase in silverweed abundance and predicted zokor damage in a warmer climate; therefore overgrazing should be prevented. However, moderate cattle grazing may be used to manage pest behaviour and damage. Because livestock do not selectively browse the monocots and dicots, they may be able to reduce the response of silverweed to warming while overgrazed grasslands are recovering, as found in the present study. Klein, Harte & Zhao (2007) have suggested using cattle gazing to halt warming-induced shrub expansion in alpine meadows of the Tibetan Plateau. To this end, proper and flexible grazing (not overgrazing) should be encouraged in a future warmer world. Finally, the complexity of trophic interactions under global warming implies that other similar ecosystems that are predicted to be more or less productive should be carefully managed to keep pest populations and damage under control. Research should address the issue of trophic interactions under global warming in a variety of different ecosystems.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

We thank Brandon Barton and Xianming Gao for insightful comments and Yibin Yuan, Jian Feng, Xinwei Wu and Junpeng Mu for field assistance. This study was funded by the action-plan for West Development (KZCX2-XB2-02) and ‘100-Talent Program’ of Chinese Academy of Sciences.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information
  • Andersen, D.C. (1987) Below-ground herbivory in natural communities: a review emphasizing fossorial animals. Quarterly Review of Biology, 62, 261286.
  • Arft, A.M., Walker, M.D., Gurevitch, J., Alatalo, J.M., Bret-Harte, M.S., Dale, M., Diemer, M., Gugerli, F., Henry, G.H.R., Jones, M.H., Hollister, R.D., Jónsdóttir, I.S., Laine, K., Lévesque, E., Marion, G.M., Molau, U., Mølgaard, P., Nordenhäll, U., Raszhivin, V., Robinson, C.H., Starr, G., Stenström, A., Stenström, M., Totland, Ø., Turner, P.L., Walker, L.J., Webber, P.J., Welker, J.M. & Wookey, P.A. (1999) Responses of tundra plants to experimental warming: meta-analysis of the international tundra experiment. Ecological Monographs, 69, 491511.
  • Báez, S., Collins, S.L., Lightfoot, D. & Koontz, T.L. (2006) Bottom-up regulation of plant community structure in an aridland ecosystem. Ecology, 87, 27462754.
  • Bakker, E.S., Knops, J.M.H., Milchunas, D.G., Ritchie, M.E. & Olff, H. (2009) Cross-site comparison of herbivore impact on nitrogen availability in grasslands: the role of plant nitrogen concentration. Oikos, 118, 16131622.
  • Barton, B.T., Beckerman, A.P. & Schmitz, O.J. (2009) Climate change affects direct and indirect interactions in an old-field food web. Ecology, 90, 23462351.
  • Barton, B.T. & Schmitz, O.J. (2009) Experimental warming transforms multiple predator effects in a grassland food web. Ecology Letters, 12, 19.
  • Chapin, F.S. III, Shaver, G.R., Giblin, A.E., Nadelhoffer, K.J. & Laundre, J.A. (1995) Responses of arctic tundra to experimental and observed changes in climate. Ecology, 76, 694711.
  • Chase, J.M., Leibold, M.A., Downing, A.L. & Shurin, J.B. (2000) The effects of productivity, herbivory, and plant species turnover in grassland food webs. Ecology, 81, 24852497.
  • Cornelissen, J.H.C., van Bodegom, P.M., Aerts, R., Callaghan, T.V., van Logtestijn, R.S.P., Alatalo, J., Chapin, F.S., Gerdol, R., Gudmundsson, J., Gwynn-Jones, D., Hartley, A.E., Hik, D.S., Hofgaard, A., Jónsdóttir, I.S., Karlsson, S., Klein, J.A., Laundre, J., Magnusson, B., Michelsen, A., Molau, U., Onipchenko, V.G., Quested, H.M., Sandvik, S.M., Schmidt, I.K., Shaver, G.R., Solheim, B., Soudzilovskaia, N.A., Stenström, A., Tolvanen, A., Totland, Ø., Wada, N., Welker, J.M., Zhao, X.Q. & Team, M.O.L. (2007) Global negative vegetation feedback to climate warming responses of leaf litter decomposition rates in cold biomes. Ecology Letters, 10, 619627.
  • Dormann, C.F. & Woodin, S.J. (2002) Climate change in the Arctic: using plant functional types in a meta-analysis of field experiments. Functional Ecology, 16, 417.
  • Duffy, J.E. (2002) Biodiversity and ecosystem function: the consumer connection. Oikos, 99, 201219.
  • Field, C.B., Behrenfeld, M.J., Randerson, J.T. & Falkowski, P. (1998) Primary production of the biosphere: integrating terrestrial and oceanic components. Science, 281, 237240.
  • Goheen, J.R., Keesing, F., Allan, B.F., Ogada, D. & Ostfeld, R.S. (2004) Net effects of large mammals on Acacia seedling survival in an African savanna. Ecology, 85, 15551561.
  • Hillebrand, H., Bennett, D. & Cadotte, M.W. (2008) The consequences of dominance: a review of the effects of evenness on local and regional ecosystem processes. Ecology, 89, 15101520.
  • Jägerbrand, A.K., Alatalo, J.M., Chrimes, D. & Molau, U. (2009) Plant community responses to 5 years of simulated climate change in meadow and heath ecosystems at a subarctic-alpine site. Oecologia, 161, 601610.
  • Klein, J.A., Harte, J. & Zhao, X.Q. (2004) Experimental warming causes large and rapid species loss, dampened by simulated grazing, on the Tibetan Plateau. Ecology Letters, 7, 11701179.
  • Klein, J.A., Harte, J. & Zhao, X.Q. (2007) Experimental warming, not grazing, decreases rangeland quality on the Tibetan Plateau. Ecological Applications, 17, 541557.
  • Lawrence, D., D’Odorico, P., Diekmann, L., DeLonge, M., Das, R. & Eaton, J. (2007) Ecological feedbacks following deforestation create the potential for a catastrophic ecosystem shift in tropical dry forest. Proceedings of the National Academy of Sciences, USA, 104, 2069620701.
  • Li, X.G., Zhang, M.L., Li, Z.T., Shi, X.M., Ma, Q.F. & Long, R.J. (2009) Dynamics of soil properties and organic carbon pool in topsoil of zokor-made mounds at an alpine site of the Qinghai-Tibetan Plateau. Biology and Fertility of Soils, 45, 865872.
  • McCarthy, J.J., Canziani, O.F., Leary, N.A., Dokken, D.J. & White, K.S. (2001) Climate Change 2001: Impacts, Adaptation and Vulnerability: Contribution of Working Group II to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge.
  • Meserve, P.L., Kelt, D.A., Milstead, W.B. & Gutiérrez, J.R. (2003) Thirteen years of shifting top-down and bottom-up control. BioScience, 53, 633646.
  • Moen, J., Gardefjell, H., Oksanen, L., Ericson, L. & Ekerholm, P. (1993) Grazing by food-limited Microtine rodents on a productive experimental plant community: does the green desert exist? Oikos, 68, 401413.
  • Muldavin, E.H., Moore, D.I., Collins, S.L., Wetherill, K.R. & Lightfoot, D.C. (2008) Aboveground net primary production dynamics in a northern Chihuahuan Desert ecosystem. Oecologia, 155, 123132.
  • Ni, J., Wang, G.H., Bai, Y.F. & Li, X.Z. (2007) Scale-dependent relationships between plant diversity and above-ground biomass in temperate grasslands, south-eastern Mongolia. Journal of Arid Environments, 68, 132142.
  • Peñuelas, J. & Filella, I. (2001) Responses to a warming world. Science, 294, 793794.
  • Polley, H.W., Wilsey, B.J. & Derner, J.D. (2007) Dominant species constrain effects of species diversity on temporal variability in biomass production of tallgrass prairie. Oikos, 116, 20442052.
  • Reynolds, J.F., Smith, D.M.S., Lambin, E.F., Turner, B.L. II, Mortimore, M., Batterbury, S.P. et al. (2007) Global desertification: building a science for dryland development. Science, 316, 847851.
  • Root, T.L., Price, J.T., Hall, K.R., Schneider, S.H., Rosenzweig, C. & Pounds, J.A. (2003) Fingerprints of global warming on wild animals and plants. Nature, 421, 5760.
  • Schlesinger, W.H., Reynolds, J.F., Cunningham, G.L., Huenneke, L.F., Jarrel, W.M., Virginia, R.A. & Whitford, W.G. (1990) Biological feedbacks in global desertification. Science, 247, 10431048.
  • Schmitz, O.J., Post, E., Burns, C.E. & Johnston, K.M. (2003) Ecosystem responses to global climate change: moving beyond colour-mapping. BioScience, 53, 11991205.
  • Solomon, S., Qin, D., Manning, M., Marquis, M., Averyt, K., Tignor, M.M.B., LeRoy, M.H. & Chen, Z. (2007) Climate Change 2007: The Physical Science Basis: Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, New York.
  • StatSoft Inc. (2000) Statistica for Windows. StatSoft, Inc., Tulsa, OK, USA.
  • Suttle, K.B., Thomsen, M.A. & Power, M.E. (2007) Species interactions reverse grassland response to change climate. Science, 315, 640642.
  • Wang, Q., Zhou, W., Wei, W., Zhang, Y. & Fan, N. (2000) The burrowing behaviour of Myospalax baileyi and its relation to soil hardness. Acta Theriologica Sinica, 20, 277283.
  • Xiang, S., Guo, R., Wu, N. & Sun, S. (2009) Current status and future prospects of Zoige Marsh in eastern Qinghai-Tibet Plateau. Ecological Engineering, 35, 553562.
  • Zhang, Y. (2000) Studies on the pattern of animal-plant interaction: the effects of plateau zokor on the biogeochemical cycling of alpine meadow ecosystem and its response to the chemical defense of plants. PhD dissertation, Institute of Zoology, Chinese Academy of Sciences, Beijing, China.
  • Zhang, K., Douglas, B.C. & Leatherman, S.P. (2004) Global warming and coastal erosion. Climate Change, 64, 4158.
  • Zhang, Y. & Liu, J. (2003) Effects of Plateau zokors (Myospalax fontanierii) on plant community and soil in an alpine meadow. Journal of Mammalogy, 84, 644651.
  • Zhang, Y., Zhang, Z. & Liu, J. (2003) Burrowing rodents as ecosystem engineers: the ecology and management of Plateau zokors Myospalax fontanierii in alpine meadow ecosystems on the Tibetan Plateau. Mammal Review, 33, 284294.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

Table S1. Temperate differences between warmed and ambient open top chambers.

Fig. S1. Monthly means of soil moisture.

Fig. S2. Daily means of relative humidity.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
JPE_1965_sm_FigS1.doc26KSupporting info item
JPE_1965_sm_FigS2.doc26KSupporting info item
JPE_1965_sm_TableS1.doc48KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.