Tussock Grassland Research Articles

Mountain beech forest succession after a fire at Mount Thomas Forest, Canterbury, New Zealand

The impact of fire on the resilience of tree and shrub species populations, and the relationships between distance from the unburnt forest margin and composition, were examined from data collected 1, 2, 4, 6, 10, and 15 years after a fire at Mount Thomas Forest in 1980. Although some scorched mountain beech trees flushed, virtually all trees died within five years. The few surviving mountain beech seedlings mostly occurred within 20 m of the forest margin and were less common on a north-facing transect than a south-facing transect. Other woody species (e.g., of Coprosma and Pseudopanax) commonly sprouted from burnt stumps. Increased density over time of saplings and seedlings >15 cm tall was primarily the result of onward growth of fire survivors, rather than post-fire establishment. Between 1981 and 1995, some of the earliest colonisers disappeared and a few herbaceous species gradually increased to become dominant. Current dominance by the exotic Agrostis capillaris appears a consequence of recent changes in nearby tussock grasslands and strongly limits beech seedling establishment. Partial canonical correspondence analysis showed that time and distance alone explained significant components of the compositional variation. Because little variation was explained by the interplay between time and distance, there was no evidence for mountain beech forest recovery as ongoing marginal spread within 15 years of the fire.

Stream faunas and ecoregions in South Island, New Zealand: do they correspond?

Water chemistry and benthic invertebrate communities of 100 headwater streams within 10 South Island, New Zealand, ecoregions were examined to assess the usefulness of ecoregions as a basis for stream classification. Water chemistry did not vary in a systematic manner among ecoregions and showed considerable variation within some of them. In contrast, invertebrate communities differed among ecore­ gions in a much more consistent way. Taxonomic richness, numbers of endemic spe­ cies, species dominance and summer faunal densities all differed among ecoregions, but little discrimination was possible using functional feeding groups. Stream faunas were generally similar between forested ecoregions, whereas the alpine (Southern Alps) ecoregion had the most distinct fauna. Geographically distinct areas that have been deforested and are dominated by tussock grassland and improved pasture now have very similar stream faunas with few regionally distinct elements present. Our results provide qualified support for the use of ecoregions as a basis for stream classi­ fication in New Zealand.

Effects of Exclosure and Management on Biomass and Soil Nutrient Pools in Seasonally Dry High Country, New Zealand

We examined the effects of grazing and three exclosure treatments (no grazing, no-grazing plus oversowing, and no grazing plus oversowing plus fertilizer) on soils (Ustochrepts) and biomass of tussock grasslands on hill country under a mean annual rainfall of about 500–600 mm in the Benmore Range, South Island, New Zealand. Excluding grazing by sheep and rabbits for 15 years, with no oversowing or fertilizer, resulted in a two- to three-fold increase in the total biomass (roots+litter+herbage) relative to the grazed treatment. On all three exclosure treatments root biomass was approximately doubled in relation to root biomass on the grazed treatment. With the exception of Mg, total nutrient pools in biomass plus soil were lowest in the grazed treatment. Oversowing alone, without added fertilizer, maintained soil pH at pre-exclosure values and prevented exchangeable Ca, Mg and K decline, but had no significant effect on soil C, N or P. In contrast, oversowing and fertilizing had the effect of increasing soil C and N by 5·7 t/ha and 0·6 t/ha respectively, relative to the ungrazed treatment, and by 8·7 t/ha and 0·87 t/ha respectively, relative to the grazed treatment. Under ungrazed, grazed, and the oversown+fertilizer treatments pH declined relative to pre-exclosure values and values on the oversown treatment. The lower pH under the oversown+fertilizer treatment was attributed to the direct and indirect acidifying effects of elemental S fertilizer, while the lower pH of the ungrazed treatment was attributed, in part, to the acidifying effect ofHieracium pilosella, with both cation removal by animals andHieraciumeffects likely to have lowered pH in the grazed treatment.A positive balance of 425 kg/ha of Ca, 680 kg/ha of K and 1900 kg/ha of N remained in total biomass plus soil on the oversown+fertilizer treatment, after fertilizer additions and previous herbage removals were taken into account. The positive Ca and K balance was attributed to release of nutrients by soil weathering, and the N balance to N fixation by legumes. Grazing of seasonally dry South Island tussock grasslands on hilly land, even at the present low levels, without fertilizing, appears incompatible with the maintenance of soil pH and carbon and total soil and biomass nutrients.

The endangered Tunbridge buttercup (Ranunculus prasinus): Ecology, conservation status and introduction to the Township Lagoon Nature Reserve, Tasmania

Abstract Ranunculus prasinus is a mat‐forming species of the marginal herbfields and tussock grasslands of brackish wetlands. It is only known from four wetlands, all on private land, in the driest part of Tasmania. The species has been successfully introduced from divisions from three of the populations into a secure reserve with similar vegetation to that of the natural populations. This is considered to be desirable and morally sound in that it increases the chances of survival of native biological diversity as a whole. However, ex situ measures and a gaining of security for natural populations would also be prudent.

Factors predisposing forests to canopy collapse in the southern Ruahine Range, New Zealand

The introduced brushtail possum Trichosurus vulpecula Kerr is recognised as the primary agent of defoliation and stand-level dieback in New Zealand broadleaved forests (except Nothofagus). The distribution and magnitude of canopy collapse of forest in 33 500 ha of the southern Ruahine Range, New Zealand was mapped from 1995 polychrome aerial photographs. Relationships between canopy collapse and forest type, altitude, aspect and slope were analysed using generalised additive models. Canopy composition was the strongest factor predicting the extent of collapse and modification: broadleaved-conifer forest is most affected, with 68–87% of the area of six such forest types replaced by scrub-low forest and tree-fernland; and surprisingly, three Nothofagus-dominated types have up to 28% of their former area now in shrubland or tussock grassland. The susceptibility to collapse of Nothofagus forest was positively correlated with distance to non- Nothofagus forest types. Subalpine scrub, which is dominated by species not favoured by possums, has increased in area by 32%, replacing former upper montane forest. Physiographic factors were much less important in multiple regressions. However, in general terms, forests on steeper slopes, in the upper montane-subalpine zone, and on warm westerly and northerly aspects were more susceptible to collapse and modification than elsewhere. It is postulated that possums were the primary agent responsible for collapse of non- Nothofagus broadleaved forest, and that red deer Cervus elaphus and goats Capra hircus were responsible for inhibiting canopy replacement by eliminating regeneration in forest understoreys. Secondary effects, such as outbreaks of defoliating insects and mechanical damage from wind also contributed to canopy collapse and were possibly triggered by possums opening up the canopy. Collapse of Nothofagus canopies possibly results from a breakdown in the recruitment phase of stand turnover, by the smothering effects of deer-induced shrubs and tussock grasses on seedlings and saplings.

The Effect of Grassland Gap Size on Establishment, Growth and Flowering of the Endangered Rutidosis leptorrhynchoides (Asteraceae)

To determine the effect of grassland gap size on patterns of seedling recruitment and juvenile establishment of the endangered composite Rutidosis leptorrhynchoides, seed and transplants were introduced into artificially created canopy gaps in both a 'short' and 'tall' temperate Themeda triandra tussock grassland in southern Victoria, Australia. Square gaps of 0 cm (undisturbed control), 15 cm, 30 cm, 50 cm and 100 cm width were created by removing the dominant grass, and the emergence, survival and growth of R. leptorrhynchoides were followed for 1 year. Survival and performance were compared with light quantity at ground level and soil moisture differences between the gaps. Emergence of seedlings was greatest in large gaps (30 cm, 50 cm and 100 cm) but survival was restricted primarily to the 100-cm gaps in both grasslands. Survival of transplants to 1 year occurred in 30-cm, 50-cm and 100-cm gaps, suggesting that juvenile plants tolerate competition for resources better than germinating seedlings do. Survival, rate of growth and total number of inflorescences produced, however, was significantly greater in 100-cm gaps. There were few differences in the seasonal pattern of soil moisture between gaps in both grasslands. The amount of light at ground level was significantly greater in the short-grassland and in large gaps (30 cm, 50 cm and 100 cm) at most times during the year. Differences between gaps in total soil moisture and light levels, however, only partly explain the patterns of transplant survival and growth observed. Soil moisture variability is suggested to have been an important factor influencing transplant survival over summer. These results confirm that R. leptorrhynchoides is a gap-sensitive species, with recruitment and survival unlikely to occur in canopy gaps less than 30-50cm in diameter. Management of remnant populations needs to ensure that large canopy gaps are regularly maintained to maximize successful seedling recruitment and maintain the standing population. This may be achieved by burning the grassland at 3-year intervals. In the absence of the frequent burning of productive grasslands, localized extinction is likely. Reintroduction of the species into secure grassland reserves is likely to be problematic given the paucity of large Themeda-free gaps in these grasslands. Alternative. stategies for conserving the species include its introduction into grassy woodlands where gap closure rates are likely to be slower.

Fog deposition in tall tussock grassland, South Island, New Zealand

Hill cloud microphysics was examined with a forward scattering spectrometer probe over tall tussock grassland at Swampy Summit near Dunedin, New Zealand. In a 4-month period, cloud water drop diameter spectra were obtained for 17 cloud–ground intersection events that varied from 1 to 61 h in duration. Rain fell in most events. Easterly conditions made up 60% of the events and had a median cloud drop diameter of 13.9 μm and 0.14 g m−3 liquid water content. For southwest winds, median drop diameter and liquid water content were significantly smaller at 9.8 μm and 0.05 g m−3.Aerodynamic conductance for momentum transfer, measured with a three-dimensional sonic anemometer, varied linearly with wind speed (R2=0.7) between 50 and 70 mm s−1 for wind speeds from 4 to 9 m s−1. Combining this relation with the cloud liquid water content measurements, cloud water deposition rates were estimated onto the 0.8-m-tall tussock canopy of one-sided leaf area index=3. These estimates were in the range 0.02–0.26 mm h−1 with a median of 0.05 mm h−1 which is the same as the median rate derived from the water balance of a large weighing lysimeter containing eight tussock plants. However, the water balance included a number of small quantities with determination of evaporation and rainfall being particularly difficult. Consequently, the lowest cloud water deposition rates for the water balance were sometimes negative. Nevertheless, even the maximum estimated rates incorporated into an annual calculation suggest cloud water deposition is unlikely to be a significant component of the water balance of a tall tussock grassland catchment. In the uplands of southern New Zealand, changing land use from tall tussock to pastoral agriculture with short vegetation is unlikely to alter streamflows via the cloud water deposition process.

Pollen-vegetation relationships on the subantarctic Auckland Islands, New Zealand

A modern pollen-vegetation cover data set of 51 sites from the subantarctic Auckland Islands is presented. The islands are completely peat covered, and extensive bogs occur throughout. Low forest and scrub covers the lowland areas of the islands, with maritime tussock and herbfield associations on exposed coasts. With increasing altitude low scrub and shrubland-grassland predominate and, above 300 m altitude, tussock grassland and fellfield. Tall, wind-pollinated species do not occur on the island, and pollen deposition is thus largely local. The broad vegetation communities are clearly recognisable by detrended correspondence analysis (DCA) of the pollen data. Although percentage cover of the major plant taxa correlates significantly in most cases with pollen percentages, reliable quantitative inferences are not possible because of the high variability. Pollen types derived from Australia and the New Zealand mainland are consistently represented in the pollen rain, averaging 0.9% and not exceeding 5%. Comparison by DCA of fossil samples from a Holocene peat core with the modern data set demonstrate a close resemblance between modern and fossil samples. This survey shows that when tall wind-pollinated trees are excluded, low-growing vegetation with poorly dispersed pollen can be accurately characterised by its pollen rain.

Hydrological impacts of converting native forests and grasslands to pine plantations, South Island, New Zealand

Changes in water yield, flood hydrology, and low flows caused by replacing indigenous forests and grasslands with commercial softwoods have been investigated in New Zealand since the mid-1970s. The long-term results of two of these studies are discussed here. The first deals with the conversion of mixed evergreen forest to pine plantation in the northwestern South Island. After a 2-year calibration period one catchment was left as the control (DC2) and the other two catchments were harvested in 1981, one by skidder (DC1) and the other by hauler (DC4), and planted in pines shortly thereafter. For the first 4 years after harvesting the average annual difference in water yields between DC1 and DC2 was 352 mm (69%), and between DC4 and DC2 it was 463 mm (90%), which equates to an annual increase of 312 mm (61%) and 344 mm (68%), respectively, when compared with the calibration period. Planting the harvested areas caused the water yield from both catchments to return to pre-harvesting levels within 8 years, and an estimated reduction in runoff of 340 mm within 5 years at DC4. Mean flood peaks increased after harvesting, especially for small and medium storms on the skidder-logged catchment (75–100%). The response of the storm quickflows to harvesting was similar but much more subdued. Low flows also increased after harvesting. Tree growth brought storm peak flows, quickflows, and low flows back to the levels of those in the original beech forest within 10 years. The second study examines the impact of converting tussock grasslands to pine plantations using data collected from two catchments in the eastern uplands of southern New Zealand. After a 3-year calibration period (1980–1982) one catchment was planted in pines over 67% of its area and the other was left in tussock. By 1989 the difference in annual water yield from the planted catchment was 130 mm, and between 1991 and 1994 it averaged 260 mm (27% of total runoff from the control). Differences in low flows (represented by the minimum annual 7-day mean) showed a similar trend, and suggest that in dry periods, afforestation of tussock grasslands can reduce water yields by 0.18 mm day −1. Higher interception losses from increased canopy evaporation is believed to be the main reason for the reduction in water yield. After 10–12 years of tree growth mean flood peaks had fallen between 55 and 65%, and quickflows had decreased by about 50%.

Comparative nutrient responses by Pinus radiata, Trifolium repens, Dactylis glomerata, and Hieracium pilosella on a Mackenzie Basin outwash plain soil

The nutrient responses by a pine (Pinus radiata), a pasture grass (cocksfoot, Dactylis glomerata), a pasture legume (white clover, Trifolium repens), and a hawkweed (mouse‐ear hawkweed, Hieracium pilosella), were compared in a pot trial using a Mackenzie Basin outwash plain soil. Responses to nitrogen (N) and phosphorus (P) were tested alone and in combination, whereas responses to sulphur (S), potassium (K), magnesium (Mg), micronutrients, and lime were tested only in the presence of both N and P. All species responded to N and P, and to S and lime in the presence of N and P, but none responded to K, Mg, or micronutrients. N was the major deficient nutrient for the non‐legume species, with grass and hawkweed responding by 700% in the presence of added P, whereas pine showed a much smaller (56%) response. Clover showed the greatest response to P. Hawkweed showed the smallest response to P, but the greatest responses to S and lime. While pine responded to the same nutrients as the herbaceous species, it is suggested that the responses by pine occurred before mycorrhizae formation, and may be more pertinent to seedlings establishing from seed in the absence of mycorrhizae, than to plants with well‐developed mycorrhizal root systems. The responses by pine to N, P, S, and lime reported here therefore need to be confirmed with plants which possess mycorrhizal root systems at the time of fertiliser application. The comparative nutrient responses by mouse‐ear hawkweed are discussed. It is suggested that hawkweed is efficient at absorbing P, an attribute that would be of competitive advantage in soils with low P availability and which may have contributed to its widespread invasion of the tussock grasslands.

Field management of the bilby Macrotis lagotis in an area of south-western Queensland

A study is reported of a population of bilbies Macrotis lagotis Reid at Davenport Downs, a 15 100 km 2 cattle property in the Channel Country of Queensland. Systematic traverses first established that bilby burrows were located throughout the tussock grasslands on downs near major watercourses on the property. A 10-month survey of burrows subsequently revealed that fewer than one-half of those studied were in permanent use. Moreover, daily burrow use in a typical burrow aggregation ranged from one-third to two-thirds of the total over a 45-day observation period. From regular observations made of 27 burrow aggregations consisting (at the end of the period) of 169 burrows, the data suggested that these may have accommodated 40 bilbies, that each added an extra burrow over 10 months, and that each used one, two or three burrows each night. Further research needs are identified from these introductory observations. These include monitoring the distribution and relationships between burrow and bilby numbers. Property management guidelines to minimize disturbance to areas of bilby activity, including by overgrazing by stock, have been recommended to, and adopted by, the landholders involved.

Growth and Vegetation History of Alpine Mires on the Old Man Range, Central Otago, New Zealand

Alpine mires are widespread on the flat-topped mountains of Central Otago, the driest and most continental region in New Zealand. Peat profiles, pollen analyses, and radiocarbon dates are presented for several mires from the Old Man Range as part of a study of the environmental history of this region. Precipitation in the early Holocene may have been up to 30% less than that of the present, and peat accumulation was restricted to topogenous mires. Alpine ombrogenous bogs began growth at ca. 7500 BP. At the same time, forest and tall scrub spread in the adjacent lowlands in response to increasing rainfall, replacing grassland-shrub-lands. Peat accumulation was slow and episodic and declined to very low rates after 3000 BP, probably as a result of cooler winters and drier summers. Low levels of natural fire occurred throughout the Holocene. However, from 750 BP onwards, lowland forest and scrub was severely reduced, subalpine shrubland declined, and snow tussock grassland increased markedly as fires lit by Polynesian hunters swept the region. From A.D. 1850 onwards, European pastoralists introduced sheep and increased the burning of the uplands to improve grazing. Peat accumulation increased in topogenous mires after burning and grazing of the alpine grassland, possibly because of accelerated runoff and nutrient input.

The relationship between land use and physicochemistry, food resources and macroinvertebrate communities in tributaries of the Taieri River, New Zealand: a hierarchically scaled approach

1. The relationship between land use and stream conditions was investigated, including physicochemistry, the availability of primary food resources and species richness, species composition and trophic structure of stream macroinvertebrate communities. The survey involved eight subcatchments of the Taieri River (New Zealand) encompassing reasonably homogeneous examples of four major land uses: native forest, native tussock grassland, plantations of introduced pine and agricultural pasture.2. Each land use was represented by two subcatchments, each subcatchment by two to four tributaries, and each tributary by two to three sampling sites. These three sampling scales each represent typical designs for stream community studies. By recording responses at all scales, it can be determined explicitly whether the scale of sampling influences interpretation of community structure.3. Elevation, riffle length, proportion of large substrata in the bed, total phosphorus and alkalinity were significantly related to land use, as were canopy cover and the relative abundance of leaves and wood in the streams. Principal components analysis of invertebrate density data identified nine orthogonal community types, the distributions of two of which were significantly related to land use. The role played by browsers and shredders in the stream community depended on land use.4. Primary analysis was at the level of the tributary. When it focused on sites within tributaries, more variables were related to land use and at a higher level of significance. This was largely a result of enhanced statistical power due to increased replication. When whole subcatchments were the focus of attention, statistical power was so low, even with six to eleven subsamples to generate overall means, that few significant patterns could be identified. However, the community patterns that were revealed were similar whatever the scale of sampling.

Benthic communities of forest streams in the South Island, New Zealand: effects of forest type and location

Algal biomass, standing stocks of particulate organic matter (POM) and benthic invertebrate communities of 20 streams draining native forest, exotic conifer forest and tussock grassland catchments in the South Island, New Zealand, were examined in spring/summer 1994-95. Sixteen were forest streams, 8 in native and 8 in exotic forest, with 4 of each kind located on each side of the Main Alpine Divide and therefore in different climatic (including rainfall) zones. Chemistry of stream water reflected geographic location more strongly than forest type with western streams being slightly more acid and having lower ionic concentrations than those in the east. Algal biomass on stones in forest streams ranged from 7-51mg chl.m -2 , and regardless of forest type, appeared to be limited primarily by light and substrate stability. POM standing stocks were low (22-254g AFDW m -2 ) and dominated by coarse materials (>13 mm) in both native and exotic forest streams. Benthic faunas of all stream groups were dominated numerically by Chironomidae and Ephemeroptera and contained similar numbers of species (35-40). However, differences in community composition were found between eastern and western streams, and these were greater than between forest types. Total invertebrate abundances were moderately low (2008- -6083 m -2 ) and were significantly correlated with epilithic algal biomass. In contrast, correlations between invertebrate numbers and detrital biomass were weak, perhaps reflecting poor retention of riparian inputs. Our results support the view that forest type has little effect on stream benthic communities in the South Island, New Zealand, and that geographical factors (climate, location) are of greater importance.

Population numbers, response to weather, movements and management of the threatened New Zealand skinks Oligosoma grande and O. otagense in tussock grassland

The threatened skinks Oligosoma grande and O. otagense currently exist in small, fragmented populations inhabiting rock outcrops among subalpine tussock grassland in southern New Zealand. This habitat is disappearing as farm production increases. Population numbers in three prime sites on Emerald Creek are low in absolute terms (11?56 individuals of each species; 15?78/ha). Low numbers and habitat fragmentation place populations at demographic and genetic risk. Emergence of both species from outcrops is highly weather-dependent, being negatively correlated with cloud cover and positively correlated with vapour pressure deficit and either air or rock temperature. Although higher numbers of both species emerge in warm, dry and sunny weather, O. grande is more likely than O. otagense to be seen in marginal conditions. Many animals showed site fidelity but some moved between outcrops; maximum distances between sightings were 139 m for O. grande and 79 m for O. otagense. Pair bonding may occur in O. otagense. We recommend that annual censuses with moderate-low confidence intervals be carried out at a range of sites, that future surveys and censuses consider weather variables explicitly, and that the effects of habitat degradation and habitat enhancement on dispersal between outcrops be investigated.

Establishment of legumes and grasses overdrilled into Hieracium ‐infested montane tussock grasslands

The establishment of lotus (Lotus corniculatus), alsike clover (Trifolium hybridum), cocksfoot (Dactylis glomerata), and mountain brome (Bromus marginatus) were compared following overdrilling by an experimental strip seeder and a conventional triple disc drill in the presence and absence of paraquat herbicide. Drilling was carried out in late August or in late January at two sites on dry, undeveloped soils in the Mackenzie Basin, New Zealand. With August sowing, the number of lotus, cocksfoot, and brome, but not alsike seedlings at the end of the first growing season, were higher with the strip seeder drill than with the triple disc drill. With January sowing, lotus, alsike, and cocksfoot seedling numbers were higher with the strip seeder drill. Numbers declined over the following four years but the advantage of the strip seeder drill remained. Improved seedling nodulation, root development, root length, and shoot length were recorded with the strip seeder drill. Drill effects on legume herbage accumulation in Years 2–4 were not consistent. There were no significant herbicide effects on seedling establishment in the first growing season but herbicide increased legume herbage accumulation in Years 2–4, particularly with the triple disc drill. The higher seedling establishment and enhanced early seedling growth indicate the potential of the strip seeder drill technology particularly for establishing species with low seedling vigour.

North Island seral tussock grasslands. 3. The influence of heather (Calluna vulgaris) on rates of change from tussock grassland to shrubland

Abstract The cover of the adventive heath Calluna vulgaris in seral tussock grassland of northern Tongariro National Park and the influence of C. vulgaris on the rates of indigenous shrub invasion of the grasslands were assessed by relating species canopy cover and height to environmental and temporal factors using generalised additive models. C. vulgaris now dominates virtually all of the area previously covered in Chionochloa rubra tussock grassland below 1200 m over 52 km2 on the northern ringplain. Only on basin floors, and the even frostier or poorly drained hollows, does C. rubra persist as a conspicuous species, but at generally less than 10% crown cover. Native shrubs are invading C. vulgaris shrubland in a process partly explicable in terms of time elapsed since the last fire, slope, topographical position, and altitude. On sideslopes at 900 m a.s.l. Leptospermum scoparium and Dracophyllum longifolium colonise and exceed the cover of C. vulgaris 38 years after burning; on crests the native shrubs take at least 60 years to dominate; and they take 75 years on basin floors. At 1100 m the times for this transformation are almost doubled. These slow rates of change, coupled with the extensive areas of early seral vegetation and the topographical dominance of crests and basin floors, point to C. vulgaris remaining a conspicuous component of ringplain landscapes for at least another 40 years. Rates of native shrub invasion of tussock grassland without C. vulgaris showed somewhat faster initial rates, but little difference after 70–80 years.

Microbial biomass, and C and N mineralization, in litter and mineral soil of adjacent montane ecosystems in a southern beech ( Nothofagus) forest and a tussock grassland

Comparisons were made of total and microbial C and N pools and C and N metabolism in litter and mineral soil of a mountain beech ( Nothofagus solandri var. cliffortioides) forest, ca. 100 m below timberline, and an adjacent tussock grassland (dominated by Chionochloa pallens), ca. 100 m above timberline, in Canterbury, New Zealand. Mean annual precipitation at the sites is ca. 1600 mm and mean annual air temperature ca. 6°C. The silt loam soils are Andic Dystrochrepts. Total C and N and microbial C and N contents in litter and mineral soil (0–50 cm depth) differed appreciably in the two ecosystems and were 1.41, 2.00, 1.71 and 1.98 times, respectively, greater on an area basis at the grassland than at the forest site. Ratios of microbial C-to-total C, microbial N-to-total N and microbial C-to-N generally declined with profile depth. CO 2C production, per unit of total C, and metabolic quotients (qCO 2 values) tended to be greater in mineral soil from the forest than from the grassland. CO 2C production, calculated on an area basis to 50 cm depth of mineral soil, was similar in both ecosystems. Net N mineralization (during 0–56 days at 25°C) was appreciable in the forest litter, but absent in tussock litter; it was similar in both systems at 0–10 cm and 20–50 cm depths of mineral soil, but was lower in the forest than in the grassland at 10–20 cm depth. Nitrification was not detected in the litter samples and was either absent or very low in the samples of mineral soil. Results, overall, show that marked differences in soil and microbial properties can occur in adjacent, indigenous ecosystems in almost the same climatic environment. Although soil microbial biomass levels were lower in the forest than in the grassland, the potential metabolic activity of the component microorganisms tended to be greater in the forest. The relevance of these results to the turnover of organic matter in these ecosystems is briefly discussed.

Effect of riparian land use on contributions of terrestrial invertebrates to streams

Since terrestrial invertebrates are often consumed by stream fishes, land-use practices that influence the input of terrestrial invertebrates to streams are predicted to have consequences for fish production. We studied the effect of riparian land-use regime on terrestrial invertebrate inputs by estimating the biomass, abundance and taxonomic richness of terrestrial invertebrate drift from 15 streams draining catchments with three different riparian land-use regimes and vegetation types: intensive grazing — exotic pasture grasses (4 streams), extensive grazing — native tussock grasses (6 streams), reserve — native forest (5 streams). Terrestrial invertebrate drift was sampled from replicated stream reaches enclosed by two 1 mm mesh drift nets that spanned the entire channel. The mean biomass of terrestrial invertebrates that entered tussock grassland (12 mg ash-free dry mass m−2 d−1) and forest streams (6 mg AFDM m−2 d−1) was not significantly different (p > 0.05). However, biomass estimated for tussock grassland and forest streams was significantly higher than biomass that entered pasture streams (1 mg AFDM m−2 d−1). Mean abundance and richness of drifting terrestrial invertebrates was not significantly different among land-use types. Winged insects contributed more biomass than wingless invertebrates to both pasture and tussock grassland streams. Winged and wingless invertebrates contributed equally to biomass entering forest streams. Land use was a useful variable explaining landscape-level patterns of terrestrial invertebrate input for New Zealand streams. Evidence from this study suggests that riparian land-use regime will have important influences on the availability of terrestrial invertebrates to stream fishes.

Aeolian dust transport and deposition by foehn winds in an alpine environment, Lake Tekapo, New Zealand

Observations from an automatic weather station and anemograph network are used in conjunction with recorded dust deposition rates to identify principle sources of wind blown dust within an alpine lake basin. Meteorological observations made throughout the study indicate that foehn windstorms provide ideal conditions for the initiation of aeolian processes, including dust storm genesis. A mean wind speed during foehn windstorms of 7.5 m s −1 at 2.65 m above the surface was observed to initiate entrainment of dust particles from geomorphically active areas within the lake basin, particularly from the dry braid channels of glacier fed rivers, exposed lacustrine deltas and degraded tussock grasslands. This threshold entrainment velocity is similar to others presented within the international literature for such surface types. A diurnal modulation of meteorological parameters monitored during foehn wind events suggests that aeolian processes, such as dust entrainment, are most likely to occur from late morning to early evening. However, where topography enhances wind speeds, for example by channelling airflow down river valleys, favourable conditions for dust entrainment may prevail throughout the night in the absence of precipitation. Potential dust roses were determined for two sites within the lake basin. The dust roses supported anecdotal and observational evidence of principle dust sources within the study area, and appear to be an effective tool for identifying the potential hazard to a particular location from blowing dust, once the physical trigger conditions for dust entrainment are known. This requires detailed and concurrent monitoring of meterological variables in association with dust transportation and deposition rates, especially in complex terrain where many micro-climates often exist.