Introduction

Wallaby grass is the common name for any of about 20 species of the genus Austrodanthonia, which are native perennial grasses common in southeast Australia. Grasslands containing Austrodanthonia spp. occupy about 1.4 million hectares in this region (Garden et al. 2000).

Field observations indicate that at least some Austrodanthonia spp. appear to be able to respond to increased fertility (Garden et al. 1996), but which species are responsive and whether responses are to phosphorus (P) from applied superphosphate, or nitrogen (N) fixed by associated legumes is unknown.

Our aim in this study was to determine the growth responses to N and P additions for a selection of eight Austrodanthonia species from the southern tablelands of NSW.

Materials and Methods

Species of Austrodanthonia were selected on the basis of being common (A. racemosa, A. pilosa, A. eriantha), productive (A. fulva, A. duttoniana), of low productivity (A. carphoides, A. auriculata), or commercially available (A. richardsonii cv. Taranna).

The experiment was conducted in a glasshouse at 25/15 ^oC day/night temperature and average daylength of 14 hours. Five plants of a species were grown in 16 cm diameter pots filled with 3 kg of sandy loam soil.

The experiment was run as two separate nutrient treatments over a range of N or P additions. Nitrogen treatments were equivalent to 0, 20, 40, 80 and 160 kg N ha^-1 applied as ammonium nitrate, half at sowing and half at 3 weeks. Phosphorus treatments were equivalent to 0, 6.6, 13.2, 26.4 and 52.8 kg P ha^-1 as triple superphosphate mixed into the top 5 cm of soil prior to sowing.

The N treatments received the high rate of P and the P treatments received the high rate of N. All pots received adequate levels of other nutrients. There were 5 replicate pots of each species for each nutrient level combination. At 6 weeks of growth, whole shoots were harvested, dried (60 ^oC) and weighed.

Mitscherlich equations were fitted to data for the relationship between level of N or P applied and shoot dry weight. The maximum growth determined from these functions was used to convert absolute growth to relative growth and Mitscherlich equations were then fitted to the relative growth data.

Critical values were determined by solving the equations for the nutrient application rate at 90% of maximum growth.

Results

The critical N requirement for the species varied over a wide range, from 26 kg N/ha for A. richardsonii to 132 kg N/ha for A. eriantha (Fig. 1). In contrast, the species critical values for P fell into two distinct groups: one with critical values averaging 18 kg P/ha and another averaging 38 kg P/ha (Fig. 2).

The rank order of species for critical nutrient requirements was generally conserved. For example, D richardsonii had low critical values and A. racemosa had high critical values for both N and P. However, there were two notable exceptions: A. eriantha which had the highest critical value for N but the lowest for P, and A. carphoides which had a low critical value for N and the highest critical value for P.

Relative growth at nil P ranged from about 25% for A. carphoides, A. duttoniana and A. pilosa to about 46% for A. fulva and A. richardsonii (Fig. 2). A. richardsonii, apparently utilises root phosphatase and citric acid exudation as mechanisms to increase P uptake from otherwise unavailable pools, and these activities are enhanced under low P conditions (Gifford et al. 1996). Whether these mechanisms operate in other Austrodanthonia spp. is unknown.

Absolute plant size ranking was an inherent species characteristic that was strongly conserved and not affected by nutrient levels (data not presented). For example, A. fulva and A. racemosa were always the largest species, and A. auriculata and A. carphoides were always the smallest species.

Fig. 1 Relative growth of Austrodanthonia spp. in response to level of applied nitrogen. Lines are the fitted functions. Symbols on the lines denote the critical nutrient level. Note that A. carphoides is hidden by A. duttoniana, and A. fulva is obscured by A. racemosa.

Fig. 2 Relative growth of Austrodanthonia spp. in response to level of applied phosphorus. Lines are the fitted functions. Symbols on the lines denote the critical nutrient level. Note that symbols for species differ from Fig. 1.

Discussion

Tall C4 grasses mainly dominated the original grasslands in southeast Australia, and Austrodanthonia spp. were probably a minor component (Garden et al. 1996).

Under sheep and cattle grazing, the composition of these grasslands has changed, and the proportion of shorter C3 grasses (including Austrodanthonia spp.) has increased.

If fertiliser is applied in addition to grazing, the proportion of Austrodanthonia spp. may increase further. Also, where grasslands have been cultivated and sown to introduced species, decline of the sown perennial grasses is often coincident with recolonisation of these areas by native perennial grasses, including Austrodanthonia spp. However, in some circumstances, grasslands dominated by introduced annual grasses (Vulpia spp., Bromus spp.) may result.

The results presented here demonstrate that there is a considerable range among Austrodanthonia spp. in their growth response to additions of N or P fertiliser. These results have implications for the competitive ability of the species in grassland communities.

At low and moderate levels of nutrient supply, species with the lowest requirement for the limiting nutrient (for example, A. richardsonii where N is most limiting and A. eriantha and A. richardsonii where P is most limiting) may be the more competitive and dominant species, as demonstrated by Wedin and Tilman (1993).

Conversely, at high levels of nutrient supply, the dominant species may be species such as A. racemosa, which have a higher requirement for nutrients but may be better competitors for light, due to inherent trade-offs in the ability to compete for nutrients and light (Tilman 1990).

This model of the mechanisms of competition agrees with our observations that A. racemosa often increases when grasslands are heavily fertilised.

We propose that this model is a useful framework for determining the driving factors controlling stability and change in the species composition of native grasslands. This understanding will allow us to determine the limits to increasing fertility (and thus productivity) of grasslands based on Austrodanthonia spp. and other native grasses without shifting these ecosystems to a eutrophic state dominated by less stable and less productive annual species.

Our results suggest that these limits will depend on the particular Austrodanthonia species present.

References

Garden, D.L., Dowling, P.M., Eddy, D. A. and Nicol, H.I. (2000). A survey of farms on the central, southern and Monaro tablelands of NSW: management practices, farmer knowledge of native grasses, and extent of native grass areas. Australian Journal of Experimental Agriculture 40, 1081-1088.

Garden, D., Jones, C., Friend, D., Mitchell, M. and Fairbrother P. (1996). Regional research on native grasses and native grass-based pastures. New Zealand Journal of Agricultural Research, 39, 471-485.

Gifford, R.M., Lutze, J.L. and Barrett, D. (1996). Global atmospheric change effects on terrestrial carbon sequestration: Exploration with a global C- and N-cycle model (CQUESTN). Plant and Soil, 187, 369-387.

Tilman, D. (1990). Constraints and tradeoffs: toward a predictive theory of competition and succession. Oikos, 58, 3-15.

Wedin, D. and Tilman, D. (1993). Competition among grasses along a nitrogen gradient: initial conditions and mechanisms of competition. Ecological Monographs, 63, 199-229.