Land Theme Report
Australia State of the Environment Report 2001 (Theme Report)
Prepared by: Ann Hamblin, Bureau of Rural Sciences, Authors
Published by CSIRO on behalf of the Department of the Environment and Heritage, 2001
ISBN 0 643 06748 5
Secondary salinity and acidity
Environmental indicators reported on in this section as originally listed and defined in Hamblin (1998):
| Environmental indicator | |
|---|---|
| L3.1 | Ratio of area of catchment under perennial annual vegetation, as proportion of total catchment (report also by State) |
| L3.3 | a | b | | Percent area of land affected by dryland salinity, and acidity, by catchment and AER |
| L3.4 | Variation in plant water utilisation with landcover change |
| L3.5 | a | b | | Index of measures to increase perennial vegetation cover, by area of catchment and AER affected |
Disruption to hydrological balance in Australia has reached a critical point for certain southern river systems, and some major artesian aquifers such as the Great Artesian Basin and Gascoyne Basin. These disruptions have occurred through direct extraction of water for irrigation and bores for stock and settlements, but also through a variety of engineering works designed to check flooding, dam water for hydroelectric power and irrigation, divert water to canal distribution systems, and channel portions of rivers. The most far-reaching effect, however, has come from changing the vegetation across very large areas. The common change is the clearing predominantly deep-rooted perennial vegetation that covered nearly all land prior to European occupation, in favour of a suite of predominantly annual crops and pastures. Some of these plants are also irrigated, often excessively. The resulting secondary or dryland salinity has become a cause of very grave concern to governments, industries and local communities in the areas affected.
Australian soils and geological strata contain very large salt stores as the result of the continent's geological history and its position in the high-pressure, arid mid-latitudes for the past 30 million years. This make Australia subject to an influx of salt from prevailing winds; today's rate is 30-100 kg/ha/year. Consequently salt stores of over 10 000 tonne/ha are found in some regions.
Watersheds define catchments, and in a landscape with sufficient relative relief catchments are the most effective management unit by which we can define geophysical integrity. In regions of very low relative relief, such as occur in many parts of Australia's interior and across extensive plains of the Darling and Cooper Basins, internal watersheds are not well defined and many rivers flow only episodically and do not reach the sea. Here catchments may be a less valuable management concept than vegetation associations.
The Murray-Darling Basin Commission and the governments of Western Australia and Victoria, together with the Commonwealth, have in recent years been particularly active in implementing a wide range of measures designed to halt (if not reverse) the trend to increasing salinisation of land and water. This issue is dealt with in greater detail in the Inland Waters Theme Report, but in this report the focus is on the relationship between vegetation cover, land use and subsequent effects on the hydrological cycle, rather than on the effect on in stream and aquifer water quality and supply.
Pressure
The proportion of land under perennial and annual vegetation in each major catchment [L Indicator 3.1]
Land managers and planners need to know how much of their regions or catchments must be in perennial vegetation to check water leaking into groundwater systems and causing more salinisation.
Most of the data on the amount of recharge that occurs under current vegetation types in agricultural landscapes have come from monitored field sites in the 300-800 rainfall regions south of latitude 30S (Ridley et al. 1997, Hume et al. 1998, Ward et al. 1999, Passioura and Ridley 1998). Comparisons with native vegetation suggest that recharge has increased from 0.1-1 mm/year to between 10 and 100 times these amounts in environments now devoted to annual crops and pastures.
Estimates of the amount of perennial vegetation needed to halt discharge of saline groundwater arising from this excess recharge vary from 30-50% of the whole of a catchment (Hatton and Nulsen 1999). This depends on the groundwater system's discharge capacity and the annual rainfall (see Table 26).
| Annual Rainfall | Groundwater discharge | Type and area of vegetation needed to halt leakage | Alteration required to present farming? | Time to slow or control leakage? |
|---|---|---|---|---|
| High >600mm | High (localised, 15-100 mm/year) | Perennial woody (tree and shrub) species over all upland areas | Yes; replant with vigorous tree species, exclude stock from all uplands and fertilise perennial pastures in valleys to produce high biomass | 10-100 years if rainfall not too high, and winters not too severe for plant growth |
|
Medium 600-400 mm |
Medium (intermediate, 5-15 mm/year) | Perennial woody vegetation over 30-50% of all land, agriculture confined to fertile productive soils | Yes; replace annual pastures with perennials (lucerne, cocksfoot), eliminate fallows, plant tree blocks at breaks of slope, and on shallow soils- exclude stock from remnant vegetation | More than 50 years, and possibly only able to slow down the process |
| Low >400 mm | Low (regional, <5 mm/year) | Perennial native vegetation over 50% of all land, and agriculture confined to fertile productive soils | Yes; needs productive, deep-rooted crops, lucerne and perennial shrubs for pastures, no fallows, replant natives in depressions and rises | Probably not possible unless most of landscape returned to native vegetation |
Sources: Walker et al. (1999) and supplementary sources.
High-rainfall environments are associated with upland regions and localised groundwater systems. Leakage far exceeds groundwater discharge capacity. In these environments it is argued that nothing short of full tree canopy cover across all parts of the landscape will check leakage sufficiently. In contrast, inland, low rainfall environments are associated with regional groundwater systems that have very slow discharge rates. Because rainfall variability is high in medium and low-rainfall regions, crops must be well adapted, and soils used should be free of gross constraints (such as very high or low pH, hard pans, high disease or pest loadings) to increase water use. Annual pastures need to be replaced by perennial browse shrubs or deep-rooted perennials such as lucerne, while all hills and depressions are returned to native vegetation.
Among scientists there are differing views on the relative effectiveness of perennial grasses and legumes, compared with trees. It has been argued that the pre-European vegetation of the Tablelands of New South Wales and Victoria was dominated by perennial grasses, and their deep and fibrous root mass provided a very much more spongy soil that increased infiltration to a greater degree than in the post-settlement era. Much of this infiltrated water would have been used where it fell through the year.
Jones (2001) has argued for an alternative conceptual model of salinity control in regions that were traditionally grass-dominated, and are still used predominantly for grazing. The Sustainable Grazing Systems program of Meat and Livestock Australia and Land and Water Australia (formally Land and Water Resources Research and Development Corporation) has also taken the approach that both introduced and native perennial pasture species, together with adequate lime and fertiliser and a well-managed system of rotational grazing, can increase water utilisation and stock productivity over much of the landscape. Tree planting may be a more localised, topical remedy but cannot realistically take the place of perennial grasslands over large areas, where stock grazing will continue to be a major land use (Sustainable Grazing Systems 1999). The key vegetational factor to all systems however, lies in the need for perennials.
Data for this important indicator have been drawn from two separate sources, and has been used to provide information at national and regional scales. In the continental treatment, the 1996-97 Agricultural Census statistics and NOAA imagery for that year were the basis to the new National Land Use Map data, (Stewart et al. 2001). This map shows the unexpected finding that there are only 10 drainage basins that have more than 50% annual vegetation, while another 30 have up to 25% annual vegetation. Fifty percent of drainage basins have 5% or less annual vegetation.
These figures seem surprising, until it is remembered that data from the last agricultural census (ABS, 1997) report that the total area of all cropping and annual pastures together is only 55 million hectares, or 7.5% of the continental land area. Native pastures, which are composed of predominantly perennial grasses with scattered woodland, account for a further 7.8%. Figure 50 very clearly demonstrates that replacement of perennials by annuals is a southern Australian problem at present. The great concern expressed by scientists about the effect of the current tree clearing in central Queensland is that this problem will inevitably spread to northern Australia if such clearing continues at today's scale and rate.
Figure 50: Perennial and annual vegetation and percentages by drainage basin.
Source: BRS (unpublished)
For land managers and local planning purposes however, the drainage basin scale is still too coarse to provide any meaningful guide to where perennial vegetation is most needed. The second method therefore focuses on mapped locations of the existing native woody vegetation, drawing on Landsat TM imagery and detailed ground- truthing, as in south-west Western Australia.
Figure 51 shows the information converted into percentage perennial cover per catchment in the agricultural belt. This information was produced for the Western Australian Salinity Action Plan by Land Monitor, a consortium of government agencies monitoring changes in vegetation and land use in Western Australia for use by landholders and the community. The map very clearly distinguishes between the tree-covered Darling Range, the coastal plain, and the cleared lands of the wheatbelt. Further east, the areas left white within the large inward-flowing catchments of the Yilgarn block indicate the boundary between the agricultural and pastoral regions. What is less easy to distinguish from this scale of catchment mapping is the difference between the heavily cleared and salted inland upper reaches of the east-west flowing rivers such as the Blackwood and Avon, and the heavily tree-covered near-coastal reaches of the same rivers. Management of upper parts of these catchments has critically affected water quality in the lower reaches.
Figure 51: Percentage perennial vegetation cover across catchments of south-west Western Australia, calculated from 1996 Landsat TM imagery.
Source: Land Monitor, WA (2000)
Implications
The scale at which data are collected, analysed and presented is crucial to the information that can be derived from it. Continental coverage assists planners and policy-makers in prioritising the areas of the country where most attention needs to focus. Where all vegetation is perennial there may be little need to invest resources into tree planting, fencing out or removing grazing animals. However, crucial information on the condition, surface cover and growth rate of the perennial vegetation that is needed to assess the capacity for native vegetation to maintain hydrological balance cannot be gleaned from these data. There are a number of regions in the rangelands where the perennial-annual vegetation mix is irrelevant for the control of leakage. The National Land and Water Resources Audit has recognised the need for such a level of detail on all vegetation systems, and has established the National Vegetation Information System (NVIS) to collect and collate data across the continent. By the next State of the Environment report in 2006 much more information on the condition and transpiration efficiency of native vegetation systems should be available.
However, continental-scale mapping and analysis will never be adequate to answer the question 'How much perennial revegetation must be established to control secondary salinity?'
The medium-scale analysis of the drainage basins, and the Western Australian example of major catchment percentages, provide more information to regional resource managers and planners in pinpointing the catchments most likely to require revegetation investment. Nevertheless, for guidance at local land manager level, detailed subcatchment maps based on actual vegetation mapping from Landsat TM (resolution 1-10m 2 ) is required, in conjunction with as much local knowledge of topography, hydrogeology and soils as can be analysed simultaneously.