Where wildflowers once bloomed across a woodland landscape, now mosaics of crops and pastures dotted with remnant trees represent the quintessential Australian countryside. Only in the rarest and tiniest of remnants can the original understorey of grassy Box woodlands still be seen.
We believe that it's important to restore some large remnants across the range of grassy White and Yellow Box woodlands to their original condition. It's important simply to augment the very few existing undisturbed remnants of this once widespread ecological community; it's important for the many species that survive only in these types of remnants; and it's important for historical reasons to show what the country was like before European settlement. Returning the native perennial groundcover can also make an important contribution to ameliorating salinity and erosion problems.
However, returning a modified woodland to what it once was is not proving easy. Much assistance has been provided for fencing of remnants, and this is especially valuable for encouraging tree and shrub regeneration. But the native grasses and wildflowers of the ground layer don't seem to recover so well. Where seed banks have been lost, some species will never return without assistance. And even then, it seems that many of the annual and perennial weeds that have come to dominate many sites do not disappear once grazing is removed.
One reason for this poor recovery may be that some fundamental conditions (such as soil nutrient levels) in the remnant have changed, so that secondary and introduced species are now favoured over the original natives. If this is the case, such changes may be need to be reversed before restoration efforts can be successful.
As a first step towards determining effective restoration methods for grassy White and Yellow Box woodlands, we set out to compare the topsoils of various degraded remnants with soils of relatively pristine remnants, to see what soil changes occur when a remnant is degraded.
We found that in ungrazed remnants, dominated by Kangaroo Grass (Themeda australis) and Snow Tussock (Poa sieberiana), soils were soft and well aerated, and favourable for plant growth in most characteristics (e.g. organic matter, total nitrogen, exchangeable cations and pH). They were low in available nitrogen, phosphorus and sulphur. Furthermore, soils beneath trees were generally higher in nutrients, especially in available phosphorus, total nitrogen, organic matter and potassium, than soils from open areas within ungrazed remnants.
Right: Grassy White Box Woodland restoration trials at 'Windermere' on the Central Western Slopes. This site is currently dominated by Red Grass (Bothrio-chloa macra) and introduced species. We aim to assist the re-establishment of Kangaroo Grass (Themeda australis), Snow Grass (Poa sieberiana) and a range of native forbs through seed and seedling reintroductions in conjunction with treatments such as hot spring burns and sugar applications that can reduce soil nitrate levels and weed seed banks.
Topsoils of modified or degraded remnants differed from those of ungrazed sites in various ways, and these differences were related to the current plant species composition of the remnants.
In partially degraded remnants dominated by Purple Wire Grass (Aristida ramosa), Wallaby Grasses (Austrodanthonia spp.) or Corkscrew Grass (Austrostipa scabra), topsoils were relatively depleted in nutrients, more acidic and more compacted than undisturbed remnants. We do not know, however, whether these changes are limiting the return of the original native dominant grasses and their associated wildflowers, or whether they simply reflect the effects of livestock grazing that simultaneously brought about the changes to the understorey.
Topsoils of other types of remnants, dominated by Red Grass (Bothriochloa macra) or Tall Spear Grass (Austrostipa bigeniculata), were generally similar to open areas of ungrazed remnants, and we suspect that these types of sites may be good starting points for low-input restoration of woodland understorey.
Soils of remnants that were almost entirely dominated by robust annual and biennial weeds such as Rye Grass (Lolium rigidum), Black Oats (Avena barbata) and Paterson's Curse (Echium plantagineum) were well aerated and relatively high in nutrients. Surprisingly though, they were comparable with soils beneath trees in ungrazed remnants for most soil characteristics, including soil phosphorus. The only striking difference between these two soils was in nitrate levels, which were extremely low in all ungrazed remnants and high in remnants dominated by annual weeds. Furthermore, soil nitrate increased proportionally with weed abundance across all types of degraded remnants in our study.
From this we suspect that soil nitrate levels are an important driver of annual weeds in grassy woodlands. We hope that lowering soil nitrate levels will help to reduce annual weeds in many remnants, and are currently trying this using techniques such as adding sugar to the soil (this temporarily ties up available nitrogen), and hot spring burns to burn off nitrogen in lush green annuals. At the same time we are adding seed of perennial native grasses, which we hope will help to maintain lower soil nitrate levels over the longer term.
We still have a lot to learn about woodland restoration, but we hope that through studies like this one, we will gain a better understanding of the processes associated with degradation, and thus will be better able to enhance the diversity of remnant woodlands in the future.
This article was first published in Woodland Wanderings, Spring 2003.