Burning in a New Zealand snow-tussock grassland: Effects on soil microbial biomass and nitrogen and phosphorus availability

Fire has been an important management tool in the pastoral use of New Zealand tussock grasslands. The effects of a farm-scale pastoral fire and subsequent grazing by sheep on soil biochemical properties in tussock grasslands dominated by the narrow-leaved snow tussock (Chionochloa rigida ssp. rigida) were investigated, 1.5 and 2.5 years after the fire event, in 0-2 cm depth mineral soil at a site at 975 m altitude in Central Otago, New Zealand. The nitrogen (N) and phosphorus (P) concentrations of C. rigida leaves were also measured. Comparisons were made with soil and tussock leaves from an adjacent unburned site. At both samplings, values of total soil organic carbon (C), extractable C, microbial biomass C, and basal respiratory activity were, on average, 14%, 18%, 23%, and 40%, respectively, lower at the burned than at the unburned site. In contrast, microbial N values were roughly similar at both sites, while microbial P values were 42% higher at the burned site after 1.5 years. Phosphomonoesterase and phosphodiesterase activities were then also similar at both sites, whereas invertase activity was higher at the burned site. The greater availability of N and P at the burned site was confirmed by the higher concentrations of N and P in C. rigida leaves sampled 2 years after the fire. Ratios of microbial C:microbial N and microbial C:microbial P were significantly lower at both samplings at the burned site, and emphasise the importance of the soil microbial biomass in conserving N and P after pastoral burning in a grassland ecosystem.     

Growth and tillering patterns within tussocks of Eriophorum vaginatum

Growth patterns were investigated for Eriophorum vaginatum tussocks from disturbed and undisturbed tussock tundra at two sites in Alaska. Total basal area of tussocks decreased with increased cryoturbation but mean basal area per tussock did not. Flowering was observed in tussocks of significantly smaller size on disturbed compared to undisturbed tundra. For tussocks with < 10% cover by shrubs and moss, number of tillers per tussock was linearly related to tussock diameter in most disturbed and undisturbed sites. Exceptions occurred in an area that had been bladed with a bulldozer 7 yr before our survey where tiller number increased as the square of diameter and in an area with much frost activity where tiller number was not emulated with diameter. The ratio of daughter tillers to adult tillers decreased with diameter in disturbed tundra, whereas the trend was less pronounced in undisturbed tundra.
Microsuccession in undisturbed tussock tundra was investigated by sampling tussocks with different amounts of shrub and moss cover for number of daughter tillers per adult tiller, weight per tiller, percent nitrogen, and percent phosphorus. Small tussocks without cover by other species and large, partially covered tussocks were not significantly different by any measures, but tussocks that were almost completely buried had significantly (p < 0.05) lower values of tillering index, weight per tiller, and percent phosphorus.     

Fire temperatures and postfire plant community dynamics in Ecuadorian grass páramo

Three aspects of the páramo vegetation's response to fires were investigated: the measurement of fire temperatures, general observations of changes in plant communities following fires, and monitoring the fate of individual plants after burning.Fire temperatures were strongly influenced by the physiognomy of the vegetation, dominated by tussocks of Calamagrostis spp. Temperatures were highest amongst the upper leaves of the tussock (sometimes >500°C). The middle levels of the tussock experienced temperatures in excess of 400°C, but in the dense leaf bases temperatures were often below 65°C. On the ground between tussocks, temperatures were variable, whereas 2 cm below ground temperatures failed to reach 65°C.Plant survival depended on the intensity of the fire and the plant's position within the tussock structure. Survival was often the result of high temperature avoidance (with buds shielded by other plant parts or buried beneath the soil surface).Post-fire Calamagrostis tiller mortality rates were high and tussock regrowth was slow. Some other species appear to maintain their populations by exploiting this recovery phase for seedling establishment on tussocks.Between tussocks, changes of occupancy at the level of the individual plants were greater after fire than in control vegetation. Most transitions were random. Those which departed from random often involved gaps and were related to post-fire mortality, regrowth from below-ground parts, colonisation or, in the case of a clonal mat-forming species, to spatial rearrangement of rosettes. Recovery was slower at higher altitude. Recovery was much slower in burned plots when the upper 2 cm of soil was removed (along with buried plant parts) compared with burned plots.Qualitative observations suggest that recovery may consist of a cyclical process, mediated by the serial dominance of several species that are physiognomically important.The frequency of fires determines the amount of fuel accumulated within grass tussocks and some plants may be unable to survive repeated burning. Chance survival of species in unburned patches of vegetation and random colonisation of gaps may be important determinants of subsequent community structure.     

Patch dynamics in grazed subtropical native pastures in south‐east Queensland

Abstract Patch formation is common in grazed grasslands but the mechanisms involved in the formation and maintenance of patches are not clear. To increase our knowledge on this subject we examined possible reasons for patch formation and the influence of management on changes between patch states in three experiments in native pasture communities in the Crows Nest district, south‐east Queensland. In these communities, small‐scale patches (tall grassland (dominated by large and medium tussock grasses), short swards (dominated by short tussock grasses and sedges), and lawns (dominated by stoloniferous and/or rhizomatous grasses)) are readily apparent. We hypothesized that the formation of short sward and lawn patches in areas of tall grassland was due to combinations of grazing and soil fertility effects. This was tested in Experiment 1 by applying a factorial combination of defoliation, nutrient application and transplants of short tussock and stoloniferous species to a uniform area of tall grassland. Total species density declined during the experiment, was lower with high nutrient applications, but was not affected by defoliation. There were significant changes in abundance of species that provided support for our hypotheses. With light defoliation and low nutrients, the tall grassland remained dominated by large tussock grasses and contained considerable amounts of forbs. With heavy defoliation, the pastures were dominated by medium tussock grasses and there were significant decreases in forbs and increases in sedges (mainly with low nutrients) and stoloniferous grasses (mainly with high nutrients). Total germinable seed densities and those of most species groups were significantly lower in the heavy defoliation than the light defoliation plots. Total soil seed numbers were not affected by nutrient application but there were fewer seeds of the erect forbs and more sedge seeds in plots with high nutrients. The use of resting from grazing and fire to manage transitions between patches was tested. In Experiment 2 , changes in species density and abundance were measured for 5 years in the three patch types with and without grazing. Experiment 3 examined the effects of fire, grazing and resting on short sward patches over 4 years. In Experiment 2 , total species density was lower in lawn than short sward or tall grassland patches, and there were more species of erect forbs than other plant groups in all patch types. The lawn patches were originally dominated by Cynodon spp. This dominance continued with grazing but in ungrazed patches the abundance of Cynodon spp. declined and that of forbs increased. In the short sward patches, dominance of short tussock grasses continued with grazing but in ungrazed plots their abundance declined while that of large tussock grasses increased. The tall grassland patches remained dominated by large and medium tussock species. In Experiment 3 , fire had no effect on species abundance. On the grazed plots the short tussock grasses remained dominant but where the plots were rested from grazing the small tussock grasses declined and the large tussock grasses increased in abundance. The slow and relatively small changes in these experiments over 4 or 5 years showed how stable the composition of these pastures is, and that rapid changes between patch types are unlikely.     

Mortality of native grasses after a summer fire in natural temperate grassland suggests ecosystem instability

The exclusion of regular fire and the introduction of livestock grazing have altered native grassland composition on Victoria's volcanic plains, commonly resulting in spear‐grass and wallaby‐grass pastures replacing Kangaroo Grass grasslands. The effect of reintroducing fire to these pastures is currently unknown, although it may be an important part of restoring this ecosystem. We measured the changes in basal area of the dominant grasses in a mixed Spear‐grass/Wallaby‐grass pastures after a summer wildfire, which we assume burnt a relatively homogenous grass sward. We found a 90–95% reduction in the basal area of live spear‐grass tussocks in burnt plots compared with unburned controls, due to the mortality of tussocks. This suggests that caution and structured experimentation should be applied when using fire to manage spear‐grass‐dominated grasslands.     

Light field heterogeneity among tussock grasses: Theoretical considerations of light harvesting and seedling establishment in tussocks and uniform tiller distributions

Although the tussock growth form of caespitose graminoids is widespread, the effect of this growth form on light interception and carbon gain of tillers has received little attention. Daily incident photosynthetic photon flux density (PFD_inc) and carbon gain in monospecific stands of tussock grasses were compared with those of a hypothetical distribution with the equivalent tiller density per total ground area, but evenly distributed rather than clumped in tussocks. This was computed for two tussock grasses Pseudoroegneria spicata (Pursh) A. Löve (bluebunch wheatgrass) and Agropyron desertorum (Fisch, ex Link) Schult. (creasted wheatgrass) at different plant densities. Daily PFD_inc and net photosynthesis (A) were greater if tillers were distributed uniformly rather than clumped in tussocks, except when the density of tussocks was so great as to approach a uniform canopy. When tussock density per ground area was low, much of the difference between tussock and uniform tiller densities in PFD_inc and A was due to shading within the tussocks; up to 50–60% of the potential carbon gain was lost in A. desertorum due to shading within tussocks. In a matrix of tussocks, the light field for establishing seedlings was very heterogeneous; potential A ranged from 7 to 96% relative to an isolated seedling. The mean of daily PFD_inc and A for seedlings in a tussock stand were nearly identical to the values in corresponding stands of uniform tiller distributions. It is hypothesized that the loss of A resulting from clumping tillers into tussocks is offset by benefits of protecting sequestered belowground resources from invasion by seedlings of competitors.     

Performance of two Picea abies (L.) Karst. stands at different stages of decline

Summary The annual replacement of tillers of Agropyron desertorum (Fisch. ex Link) Schult., a grazing-tolerant, Eurasian tussock grass, was examined in the field following cattle grazing. Heavy grazing before internode (culm) elongation seldom affected tiller replacement. Heavy grazing during or after internode elongation, which elevates apical meristems, increased overwinter mortality of fall-produced tillers and reduced the number and heights of these replacement tillers. Unexpectedly, tussocks grazed twice within the spring growing season tended to have lower overwinter tiller mortality, greater tiller replacement, and larger replacement tillers than tussocks grazed only once in late spring. These responses of twice-grazed tussocks, however, were still less than those of ungrazed tussocks or tussocks grazed moderately in early spring. The presence of ungrazed tillers on partially grazed tussoks did not increase the replacement of associated grazed tillers relative to tillers on uniformly grazed plants. This result indicates that resource sharing among tillers, if present, is short-lived or ecologically unimportant in this species. Although A. desertorum is considered grazing-tolerant, tiller replacement on heavily grazed tussocks, particularly those grazed during or after internode elongation when apical meristems were removed, was usually inadequate for tussock maintenance. These observations at the tiller (ramet) level of organization in individual tussocks (genet) may explain the often noted reduction in stand (population) longevity with consistent heavy grazing.     

Shrub succession and invasibility in a New Zealand montane grassland

Abstract Two successive shrub invasions of a short tussock grassland induced by grazing and burning were examined in montane South Island, New Zealand. The first invasion was by a native shrub, matagouri (Discaria toumatou Raoul). The second invasion was by an exotic shrub, Scotch broom (Cytisus scoparius (L.) Link), which invaded the matagouri shrubland that had developed over the grassland. The invasions were investigated using analysis of spatial patterns of both shrubs and tussocks, and age, growth rates and size structure of the shrubs. Competition between the two shrub species was examined using spatial patterns and comparing allometric relationships. After initial invasion by matagouri of the grasslands, stand density increased by consolidation about its initial colonization points. Current matagouri distribution is often negatively associated with tussocks. Scotch broom occurs most frequently in a dense sward of introduced grasses and occasionally in tussocks in interstices among matagouri shrubs. Despite the palatability of Scotch broom to sheep that graze the site, there was no evidence that the spiny matagouri facilitates invasion by protecting Scotch broom seedlings; rather there was negative association between the shrub species. The two species probably compete for above-ground space. However, diameter and height growth rates of Scotch broom far exceed those of matagouri so Scotch broom is likely to increase in biomass rapidly at the site. The autogenic organization and disturbance history of the resident plant communities have rendered each vulnerable to successive invasions.     

Changes in vegetation states in grazed and ungrazed Mackenzie Basin grasslands, New Zealand, 1990-2000

Changes in vegetation from 1990 to 2000 were examined at 10 high country localities, representing four grassland types: fescue tussock (Festuca novae-zelandiae), snow tussock (Chionochloa rigida), red tussock (C. rubra), and silver tussock (Poa cita). At each locality, three treatments were established: ambient sheep+rabbit grazing, rabbit grazing only, and no grazing. The mutivariate methods of classification and ordination were used on individual-quadrat cover data to define vegetation states and to examine transitions between them over time. Vegetation states in quadrats already dominated by Hieracium pilosella(> 50% cover) in 1990 showed little change in species composition regardless of grassland type and grazing treatment. In fescue tussock grassland, H. pilosellaincreased regardless of grazing treatment in states with low initial H. pilosellacover (< 5%), while the cover of Carex colensoi, Aira caryophyllea and Rumex acetosella decreased. In the single silver tussock locality, Poa citadecreased markedly in the ungrazed treatment as adventive species such as Dactylis glomerataand Echium vulgare increased. However, Poa citaalso decreased, probably due to drought, in the grazed treatment. Snow tussock and red tussock grassland states were more stable than those in short tussock grasslands, but there was also a general trend towards increasing H. pilosellacover in intertussock vegetation regardless of treatment. However, at one snow tussock locality, transitions from H. piloselladominated to C. rigida-dominated states occurred in ungrazed quadrats, while the reverse occurred in grazed vegetation. Implications for the management of tussock grasslands for conservation are discussed.     

Post‐fire shrub recruitment in a semi‐arid grassland: the role of microsites

Question: (1) Which factors regulate post‐fire recruitment and spread of the shrub Senecio bracteolatus in Patagonian grasslands? (2) What is the role of the grass Stipa speciosa on S. bracteolatus establishment in the post‐fire succession? Location: Northwest Patagonia, Argentina. Methods: We studied the effect of fire on S. bracteolatus recruitment and density by comparing these variables between burned and unburned grasslands. In burned areas, we compared abiotic characteristics and seedling establishment under the canopy of grasses (S. speciosa) and in gaps (inter‐tussock areas). Post‐fire interactions between S. bracteolatus seedlings and S. speciosa were studied using field and greenhouse experiments. Results: Density of S. bracteolatus was higher in burned than in unburned areas. In burned sites, seedlings were more abundant under tussock grasses, whereas juveniles were more abundant in gaps. Tussocks generated more attenuated micro‐environmental conditions than gaps during stressful summers. Gaps were more abundant in burned sites, while “under tussock” microsites were more frequent in unburned sites. In burned areas, tussocks allowed higher establishment of seedlings (facilitation), but gaps allowed more seedling growth and higher persistence of juveniles. Conclusions: Fire promoted S. bracteolatus recruitment in Patagonian grasslands by increasing the availability of favourable gap microsites. Grass protection for shrub seedlings became negative with time, probably due to competition with grasses. Gaps led to better performance and persistence of shrub plants. Six years after fire, higher shrub recruitment and adult density (observed as a trend) in burned grassland provides an opportunity for potential S. bracteolatus invasion.     

Vegetative structure,microclimate, and leaf growth of a páramo tussock grass species,in undisturbed,burned and grazed conditions

In neotropical alpine grasslands (páramo), the natural tussock grass vegetation is extensively grazed and occasionally burned. The low productivity of the tussock grass seems to be the reason for the disappearance of this growth form in the most frequently intervened areas. The structure, microclimate and leaf elongation rates of new emerging leaves were studied for the dominant tussock grass species Calamagrostis effusa, at an undisturbed, a moderately grazed (7 year after fire) and a heavily grazed (3.5 years after fire) site. In absence of grazing and burning, the tussocks had a high standing crop (1.07±0.09 kg DW · m^-2) and leaf area per projected tussock cover (LAI: 9.6±1.4). Two thirds of the total mass was dead and more than half of the leaves were in horizontal position. The tussock growth form protects the meristems from severe climatic conditions. At midday, the temperature was higher at meristem level than in the rest of the tussock. At this level, photosynthetic irradiance (PI) was almost extinct at 2.9±0.74% of PI above the vegetation. The red/far red ratio (R/FR) was strongly decreased. Initial leaf elongation of new born leaves was 2.3 mm · day^-1, and constant during the year; estimated net annual production was 198±73.8 g m^-2. At the moderately grazed and the heavily grazed study sites, the tussocks were smaller, greener and more erect than those at the undisturbed site. More PI reached the meristems and R/FR was higher at the base of grazed tussocks. Leaf elongation rates were lower. Most of the litter disappeared during the fires. The lower elongation rate of leaves in the grazed areas might be a response to defoliation, resulting in increased tillering and a lack growth associated with poor temperature insulation and more UV-B damage.     

Big and aerial invaders: dominance of exotic spiders in burned New Zealand tussock grasslands

As post-disturbance community response depends on the characteristics of the ecosystem and the species composition, so does the invasion of exotic species rely on their suitability to the new environment. Here, we test two hypotheses: exotic spider species dominate the community after burning; and two traits are prevalent for their colonisation ability: ballooning and body size, the latter being correlated with their dispersal ability. We established spring burn, summer burn and unburned experimental plots in a New Zealand tussock grassland area and collected annual samples 3 and 4 years before and after the burning, respectively. Exotic spider abundance increased in the two burn treatments, driven by an increase in Linyphiidae. Indicator analysis showed that exotic and native species characterised burned and unburned plots, respectively. Generalised linear mixed-effects models indicated that ballooning had a positive effect on the post-burning establishment (density) of spiders in summer burn plots but not in spring plots. Body size had a positive effect on colonisation and establishment. The ability to balloon may partly explain the dominance of exotic Linyphiidae species. Larger spiders are better at moving into and colonising burned sites probably because of their ability to travel longer distances over land. Native species showed a low resilience to burning, and although confirmation requires longer-term data, our findings suggest that frequent fires could cause long lasting damage to the native spider fauna of tussock grasslands, and we propose limiting the use of fire to essential situations.     

火烧对长期封育草地土壤碳固持效应的影响

火烧是内蒙古典型草原的自然现象,对长期封育草地碳固持效应具有潜在的重要影响。基于野外控制火烧实验(未火烧、1a火烧1次、2a火烧1次、4a火烧1次),采用土壤和土壤粒级组分相结合的指标体系,分析了火烧以及火烧频率对长期封育草地土壤碳贮量的影响。结果表明:火烧会降低长期封育草地0—30 cm土壤碳贮量;其中,频繁火烧将显著降低草地土壤碳贮量,而4a 1次的火烧对土壤碳贮量影响较小。火烧对0—10 cm土壤碳贮量影响明显,而对10—30 cm土层影响较小。此外,火烧对长期封育草地土壤砂粒和粉粒碳氮贮量影响较大、对粘粒碳氮贮量影响较小。火烧后表层土壤砂粒和粉粒C∶N比下降,表层土壤有机质的稳定性有所提高。与自由放牧草地相比,连续遭受火烧处理的长期封育草地仍具有较高的碳贮量。总之,火烧会一定程度降低长期封育草地的碳贮量,但并不会彻底改变其显著的碳固持效应。     

Population age structure and recent

Mid to late 20th century expansion of Dracophyllum scrub into tussock grassland on subantarctic Campbell Island has been attributed to the collective effects of global warming, cessation of farming in 1931, and continued grazing by feral sheep. To determine the importance of these, we dated the timing of scrub expansion by aging 241 Dracophyllum plants in 17 plots chosen to sample the range of environments this shrub/ small tree occupies on Campbell Island. Three plots, in lowland, well-drained, locations were dominated by large, old Dracophyllum that had established between 1846 and 1940. Virtually all shrubs in the remaining plots had established after 1940, with peaks in 1970 and 1985. The pattern of establishment does not coincide with any marked change in the temperature regime, although a prolonged period of relatively dry winters (c. 1970–1990) coincides with a late surge of regeneration on very wet sites. The removal of feral sheep from different parts of the island at different times is also unrelated to the pattern of establishment. In contrast, Dracophyllum spread follows farm abandonment in 1931 when regular burning ceased, suggesting that Dracophyllum is invading sites from which it was excluded by fire. However, the earliest reports from the mid 19th century indicate that tussock grassland was previously the dominant vegetation cover on the island, with limited Dracophyllum scrub. It is possible that the reduction of the dense tussock grasslands by fire and grazing in late 19th–early 20th century opened a regeneration window for Dracophyllum scrub to spread once burning ceased.     

Negative Plant–Soil Feedback and Positive Species Interaction in a Herbaceous Plant Community

Increasing evidence shows that facilitative interaction and negative plant–soil feedback are driving factors of plant population dynamics and community processes. We studied the intensity and the relative impact of negative feedback on clonal growth and seed germination of Scirpus holoschoenus, a ‘ring’ forming sedge dominant in grazed grassland, and the consequences for species coexistence. The structure of aboveground tussocks was described. A Lithium tracer assessed belowground distribution of functional roots. Seed rain and seedling emergence were compared for different positions in relation to Scirpus tussocks. Soil bioassays were used to compare growth on soil taken from inside and outside Scirpus tussocks of four coexisting species (Mentha acquatica, Pulicaria dysenterica, Scirpus holoschoenus and Dittrichia viscosa). We also compared plant performance of dominant plant species inside and outside Scirpus tussocks in the field. The ‘ring’ shaped tussocks of S. holoschoenus were generated by centrifugal rhizome development. Roots were functional and abundant under the tillers and extending outside the tussocks. The large roots mats that were present in the inner tussock zone were almost all dead. Seedling emergence and growth both showed a strong negative feedback of Scirpus in the inner tussock zone. Scirpus clonal development strongly reduced grass biomass. In the degenerated tussock zone, Pulicaria and Mentha mortality was lower, and biomass of individual plants and seed production were higher. This positive indirect interaction could be related to species-specific affinity to soil conditions generated by Scirpus, and interspecific competitive release in the degenerated tussock zone. We conclude that Scirpus negative feedback affects its seedling emergence and growth contributing to the development of the degenerated inner tussock zone. Moreover, this enhances species coexistence through facilitative interaction because the colonization of the inner tussock zone is highly species-specific.     

Monk's tonsure-like gaps in the tussock grass Spartina argentinensis (Gramineae).

Monk's tonsure-like gaps develop inside gramineans and other plants. The tonsures of Spartina argentinensis originate as a result of tussock development and disturbance. As the tonsure develops the ring of tillers around it breaks down and new tussocks develop from the fragments, regenerating the grassland matrix vegetatively. The microenvironment inside the tonsure is different from the surroundings and microhabitat-specific taxa grow there.     

Fire frequency regulates tussock grass composition,structure and resilience in endangered temperate woodlands

Abstract The importance of disturbance for regulating the structure and diversity of grassy ecosystems is widely recognized, but disturbance‐mediated interactions between grassland composition and grassland resilience, and consequent implications for conservation management, are less well documented. We established replicated burning, mowing and (non‐livestock) grazing regimes in two contrasting grassy woodland remnants in south‐eastern Australia, and monitored the dynamics and resilience of the matrix‐forming tussock grasses, Poa sieberiana (Poa) and Themeda australis (Themeda), over 12 years. Introduction of frequent burning to a Poa‐dominated understorey in a rarely burnt woodland enhanced dominance by Themeda, and conversely, reduced fire frequency in a frequently burnt Themeda grassland substantially increased Poa abundance. Burning was potentially detrimental in the Poa‐dominated woodland, but sward resilience (recovery after the 2002 burn) increased as Themeda increased with repeated burning. By contrast, the Themeda grassland was resilient to 4‐ and 8‐yearly burning, but biennial burning led to poor resilience and high tussock mortality under drought conditions. Contrary to other mesic grasslands, cessation of burning had not caused sward collapse by 14 years post‐fire despite high litter accumulation, potentially due to compensatory growth of Poa, lower site productivity and drought. Biennial mowing without slash removal was similar to 4‐yearly burning in effects, while exclusion from kangaroo and rabbit grazing significantly increased sward biomass and contributed to increased Poa cover and inflorescence production. We conclude that functional complementarity associated with mixed dominants enhances resilience to variable disturbance regimes, and that below certain thresholds of abundance of each dominant, this resilience declines. Conservation management of Themeda–Poa ecosystems should thus aim to maintain an effective balance of these dominants.     

火烧对羊草草原植物群落组成的影响

在羊草草原正常能着火季节内,通过人为点烧的方法,对比研究了不同时间火烧对植物群落组成的影响．结果表明,羊草－杂类草草原早春火烧后,群落密度、种类丰富度和多样性提高,均匀性降低;羊草典型草原秋季火烧后,群落密度、多样性和均匀性降低,种类丰富度提高．连续２次火烧除羊草和几种１—２年生植物密度增加外,其它各种群密度降低,群落多样性也降低,一些种类退出群落．     

Kaimanawa Feral Horses and Their Environmental Impacts

Feral horses (Equus caballus L.) occupy 64 000 ha of montane- subalpine tussock grassland in the south-western Kaimanawa Mountains, an area zoned for military training. Since 1979, the population has increased at 16.7% per annum, reaching 1102 in 1990. The most extensive habitat, red tussock (Chionochloa rubra) grassland, was variably affected by horses; tussocks in restricted mesic sites were heavily grazed and mostly eliminated, but those in extensive xeric grasslands showed little impact. The mixed hard tussock (Festuca novaezelandiae)/red tussock grasslands on basin floors and plateaux, which had already been degraded by early European farming, were suffering further depletion from horse grazing. The restricted, high altitude Chionochloa pallens tussock communities were being eliminated rapidly through preferential grazing. Oligotrophic bogs, on the summits and basin floors were largely intact, whereas high nutrient flushes were severely affected by trampling and grazing. Horses appeared to have had little impact upon Nothofagus forest understoreys. Ten plant species, several of which are vulnerable nationally, occur in the North Island only within the wild horse range. The habitats of five of them were damaged by horses. Throughout the wide basins and plateaux of the north, horses compromised floristic, rare plant habitat, and landscape nature conservation values. Their numbers may therefore have to be controlled.     

Kobresia tibetica tussocks facilitate plant species inside them and increase diversity and reproduction

Many plant species grow inside tussocks of some graminoids, but the underlying mechanisms remain unclear. We address whether some species occur and flower mainly inside tussocks so that species diversity and sexual reproduction are higher inside than outside tussocks, and whether relieving biological and physical stress is the mechanism associated with the facilitative process. In a heavily grazed grassland on the eastern Tibetan Plateau, where both physical (due to high altitude) and biological conditions (due to heavy grazing) are extremely harsh, we investigated vegetation in paired plots inside and outside 150 Kobresia tibetica tussocks and measured tussock basal area (=plot area). We also measured temperatures at soil surface, 5 and 10 cm depth and the number of animals (yaks, sheep and horses) grazing inside and outside tussocks. Sixty-seven percent of the species occurred and 42% flowered more frequently inside than outside tussocks, but none less frequently. Inside tussocks 78% species flowered, but outside tussocks only 31% did. Consequently, number of species, number of flowering species and number of inflorescences were all markedly larger inside than outside tussocks. Differences in number of species, number of flowering species and number of inflorescences inside and outside tussocks increased with increasing tussock basal area. Soil temperatures were lower inside than outside tussocks, but grazing intensity was much larger outside tussocks. Therefore, tussocks of K. tibetica facilitated the species inside them likely by grazing prevention, but not by increasing warmth. This study provides evidence that plant species colonizing tussocks of graminoids can be facilitated by the tussock species, and facilitation by grazing prevention may be one mechanism causing the coexistence of the species inside tussocks.