• environmental trend;
  • grazing gradients;
  • monitoring;
  • non-equilibrium rangelands;
  • remote sensing


  1. Top of page
  2. Abstract
  3. Introduction
  4. Development of methods
  5. Application to specific cases
  6. Conclusions
  7. References

1. Change in environmental conditions in the complex non-equilibrium rangelands of arid Australia is difficult to monitor. We show how trends in rangeland condition can be identified from changes over time in the pattern of vegetation growth across gradients of differing grazing intensity.

2. Grazing intensity was measured indirectly using distance from water. Vegetation growth was derived from remotely sensed vegetation index values before and after large rainfalls. The amount of growth was adjusted for initial vegetation cover to give a standard measure of vegetation response.

3. A vegetation response ratio was derived by comparing areas less than 4 km from water with benchmark areas further away. Systematic changes in this ratio over time indicate a trend.

4. Ratio values from test areas suggested decline, improvement and no change, consistent with recent management history.

5. The method can be applied where the whole area is affected by grazing and relatively pristine benchmarks are unavailable. It could therefore be useful in the semi-arid rangelands where paddocks are smaller than in the arid part of Australia. It also has possible uses in the rangelands of Africa and the Americas. There is potential for applying the method to traditional grazing systems as well as to commercial pastoralism.

6. The method is cheaper and more effective than other techniques and increases the capacity of grazing gradient-based monitoring schemes for arid and semi-arid areas.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Development of methods
  5. Application to specific cases
  6. Conclusions
  7. References

Degradation of drylands (the bulk of which is used as rangeland) has occurred on a global scale, with more than 70% of the area affected in Africa, Asia and the Americas and about 54% in Australia (Cardy 1994). However, these figures are only estimates and the true extent and continuing nature of the problem have proved difficult to quantify. This is because many rangeland areas show non-equilibrium behaviour (DeAngelis & Waterhouse 1987) in which short-term rainfall variability imposes dramatic changes in vegetation cover that mask any downward trend in condition except in the most extreme cases (e.g. Foran, Bastin & Shaw 1986). This lack of precision has resulted in controversy about the status and trend of rangelands world-wide, particularly in the African Sahel, where the debate continues on whether the environmental changes labelled as ‘desertification’ represent massive land degradation (Lamprey 1988) or a misinterpretation of climatic variability (Hellden 1991). Non-equilibrium behaviour also makes it difficult to determine whether land is continuing to degrade, remaining in a stable condition, or improving, even where data from remote sensing satellites showing change over time are available.

Australia is a case in point. While large areas of its grazed arid and semi-arid areas are known to be partially degraded (e.g. Tothill & Gillies 1992; ANZECC/ARMCANZ Joint Working Group 1996), rangeland monitoring and assessment schemes using a variety of methodologies have had little success in determining whether adverse changes are continuing (Graham et al. 1990; Graetz, Fisher & Wilson 1992; Department of Environment, Sport & Territories 1996; Duckett et al. 1996). This situation has arisen not only because of the difficulties of interpreting non-equilibrium behaviour, but also because vegetation response to rainfall varies in space, as does the pattern and impact of grazing. This variability, when coupled with the vast scale of grazing operations (individual enterprises may cover hundreds or even thousands of km2), makes it difficult to devise effective sampling and measurement schemes that separate trend from short-term changes (e.g. Pickup 1989; Friedel, Pickup & Nelson 1993; Jessup, Andrew & Lay 1994). The result has been claims and counterclaims about continuing degradation in the rangelands (e.g. Condon 1986; Palmer 1991; Pickard 1993) and a reluctance on the part of regulatory bodies to take action in all but the most obvious cases.

This paper describes a method for identifying trends in the state of arid and semi-arid rangelands over one or two decades by applying grazing gradient methods to a time series of remotely sensed data from the Landsat Multispectral Scanner (MSS). It complements previous studies that describe how vegetation cover can be estimated from MSS data (Pickup, Chewings & Nelson 1993), how grazing produces patterns of cover change in time and space (Pickup & Chewings 1988), and how those patterns can be used to measure the extent of degradation at a particular time on a repeatable and reliable basis (Bastin et al. 1993; Bastin, Sparrow & Pearce 1993; Pickup, Bastin & Chewings 1994). The method has potential for use in arid and semi-arid rangelands wherever there is large-scale grazing centred on fixed water points, including parts of Africa, Asia, the Americas and Australia. Some of the principles might also be adapted to other grazing systems where there are distinctive and detectable spatial patterns of grazing impact dominated by factors other than the location of water (e.g. Senft et al. 1987).

We begin with a brief description of the spatial filtering techniques used in the grazing gradient method. We then examine how vegetation cover changes with time and describe how vegetation response to rainfall may be used as a variable in the filtering procedure to identify trends in the level of degradation or recovery over time. We apply the method to examples from the rangelands of central Australia where changes are occurring because of particular land management regimes. These show that trends in condition can be identified and that the method can be used in operational monitoring programmes. The method also has potential as an early warning technique if applied routinely.

Development of methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Development of methods
  5. Application to specific cases
  6. Conclusions
  7. References

Spatial variability and the grazing gradient method

A whole suite of methods for rangeland condition assessment has been developed using data on plant species composition (e.g. Holm, Burnside & Mitchell 1987; Bastin 1989). However, the only data set with the potential to describe rangeland conditions over large areas, which is widely and regularly available at a reasonable cost, comes from high resolution sensors on remote sensing satellites such as Landsat MSS or TM (Thematic Mapper). These data can be transformed into vegetation indices that are closely related to vegetation cover (e.g. Graetz, Pech & Davis 1988).

Although there is a well-established link between lack of vegetation cover and various forms of land degradation (e.g. Frank 1984; Warren & Hutchinson 1984; Eldridge & Rothon 1992), a reduction in cover is not, in itself, sufficient grounds to indicate a decline in land condition. Indeed, most changes in cover occur because of the short-term rainfall variability characteristic of arid and semi-arid environments. It is therefore necessary to have methods that separate grazing effects from natural change and that distinguish long-term grazing impact from that which is short-term. This leads to an approach to land degradation assessment based on loss of resilience, here defined as a reduction in the ability of a landscape to recover after it has changed (Holling 1973). This reduction arises from processes such as soil erosion, reduction in the infiltration or moisture-holding capacity of soils, loss of seed banks, and increases in unpalatable woody shrubs to the extent that herbage growth is limited (Pickup, Bastin & Chewings 1994). These processes are difficult to reverse in most landscape types and recovery from them may require from decades to centuries, even if grazing is removed.

Plant cover not only changes through time with rainfall, it also varies in space, both as a function of grazing and as a result of natural variability. The principal sources of natural variability are differences in geology, soils and geomorphic history, expressed as land units or land systems; surficial patterns of run-off, run-on, erosion and deposition expressed as erosion cell mosaics; and history of burning (Pickup 1989; Stafford Smith & Pickup 1990). In arid and semi-arid Australia, most of the spatial variability due to grazing occurs because animals are confined by fences and must rely on wells or dams for drinking water. Animal movement to and from these watering points produces radial patterns in uniform country, with grazing impact decreasing with distance from water. In non-uniform country, the radial patterns are distorted into star shapes with axes of concentration extending into those landscape types that are more preferred by animals due to more palatable forage (Pickup & Chewings 1988).

Grazing gradient methods use the spatial pattern produced by grazing animals as a spatial filter to separate the impact of grazing on vegetation cover or cover change over time from that of other factors. The methods are implemented in a Geographic Information System (GIS) environment and, in this paper, use the PD54 vegetation index derived from Landsat MSS data as a measure of cover (Pickup, Chewings & Nelson 1993). The index may also be derived from Landsat TM data (Bastin, Chewings & Pearce 1996). Essentially the index uses the data space that occurs when reflected radiance data in the green and visible red spectral bands are plotted against each other. The upper limit of this data space usually indicates bare soil, while the lower area is characteristic of areas with 100% vegetation cover. Intervening points are closely correlated with the amount of vegetation cover present and can be scaled to reduce or remove differences in brightness or greenness of ground cover. The GIS used in this study covers a 60 000 km2 area of central Australia and contains vegetation index data at 1 ha resolution spanning every major vegetation growth pulse between 1982 and 1995 as well as some earlier data. These data were acquired 6–8 weeks after major rainfalls when cover was close to the maximum, during intervening droughts when cover was at its lowest, and at intervening times. The GIS also contains information on the location of water points, fence lines and other barriers to the movement of grazing animals (e.g. mountain ranges), and a stratification of the area into different landscape types based on supervised classification of MSS data (see Bastin et al. 1993 for details). The filtering is carried out by calculating the distance from water in the GIS, dividing the distance into discrete classes, and determining mean vegetation cover or cover change in each class after stratifying by landscape type. The result can then be plotted as a graph of cover against distance from water.

Grazing gradient analyses produce different information over time. For example, changes in the shape of the cover–distance from water curve, as the landscape moves into drought and forage is depleted over time by grazing, have been used to infer animal distributions (Pickup & Chewings 1988; Cridland & Stafford Smith 1993) and differences in forage palatability (Pickup 1994). Alternatively, the shape of the curve after a major vegetation growth period can be used as an indicator of the extent of degradation, while the pattern of change from before, to the height of the growth period, shows the type of degradation (Pickup, Bastin & Chewings 1994). Until now, however, there has been no method to determine whether the amount of degradation is increasing or the landscape is recovering from past grazing management practices.

Using patterns of change through time

Plant growth in arid and semi-arid areas occurs as a series of pulses followed by periods of decline as vegetation is consumed by grazing and lost by natural decay. While the size of these pulses depends on the amount of rainfall and the pre-existing cover (e.g. Hacker et al. 1991; Hobbs, Sparrow & Landsberg 1994; Pickup 1995), it also varies with land condition. Indeed, Bastin et al. (1993) have defined rangeland degradation as a grazing-induced long-term reduction in the ability of a landscape to respond to rainfall. Other things being equal, changes in pulse magnitude through time should therefore indicate changes in landscape resilience and, if they vary systematically along grazing gradients, measure grazing-induced degradation or recovery from it. This principle provides the key to degradation assessment and the identification of trend. The next step is to turn it into a practical method given the limitations imposed by poor rainfall data and the limited number of growth pulses in the archived remote sensing data record.

There is some argument about the time scale over which landscapes are affected by, and recover from, grazing in environments characterized by episodic rainfall and short-term climatic variability (e.g. Condon 1986; Hayes 1987; Perkins & Thomas 1993; Ellis, Coughenour & Swift 1993). It is therefore important to know whether a particular sequence of growth pulses has the ability to reflect a real trend in land conditions or merely short-term variation.

A typical pattern of plant cover change through time for central Australia is shown in Fig. 1 and illustrates the variability problem. On a broad scale, there are three long periods of below-average growth resulting from major droughts starting in the late 1890s, the late 1920s and the early 1960s, and two exceptional growth pulses, one during 1920–21 and the other during 1973–75. These pulses resulted from rainfalls three times greater than the annual average (260 mm) sustained over 1–2-year periods. However, most vegetation cover change occurs with greater frequency and in response to smaller rainfall events in which 50–150% of the annual average falls in a period of a few days to several months. There were five of these events in central Australia between 1982 and 1995 and several periods of relatively minor drought. This is reasonably typical of, or perhaps slightly better than, the normal run of conditions in the 120-year record.


Figure 1. Modelled vegetation biomass in central Australia from 1880 to 1995. The data were generated from rainfall observations for Alice Springs using the model developed by Pickup (1995, 1996) and show herbage biomass on a 500-km2 paddock of nearby Owen Springs Station occupied by mixed ephemeral and perennial grasslands, shrub-dominated alluvial fans, and eucalypt and acacia woodlands.

Download figure to PowerPoint

The major growth pulses in the early 1920s and 1970s had a substantial impact on the vegetation of central Australia. Anecdotal evidence, air photographs and some observations (Foran et al. 1982; Friedel 1985; Griffin & Friedel 1985; Cunningham 1996) indicate that major recruitment of tree and woody shrub species occurred. Herbage probably also became established in areas where it had previously been lost due to grazing or during drought (Friedel 1984; Purvis 1986). Little is known about the 1920–21 pulse but comparisons of satellite imagery from 1972 and the early 1980s suggests that this was the case for the 1973–75 pulse. Even so, most long-term grazing gradients did not disappear.

While some authors have argued that the appearance of a landscape after major rainfall pulses provides the most accurate assessment of its state (e.g. Condon 1986; Hayes 1987; Palmer 1991), we suggest that assessment on this basis is skewed towards unusual climatic situations and could mask underlying trends in land condition related to anthropogenic effects. It also seems unrealistic to use growth events with a 50-year recurrence interval when most vegetation growth depends on rainfall events of a much higher frequency. We therefore argue that trends in range conditions should be defined as a consistent pattern of change in the ability of a landscape to respond to rainfall prevailing over one or two decades. Thus, the five growth pulses in the remote sensing record between 1982 and 1995 should be sufficient to identify such a pattern.

Trend detection procedure

Three problems must be overcome to detect systematic change over time in a series of growth pulses. First, the effect of initial vegetation cover on the size of the growth pulse must be removed. Secondly, the five pulses in the remote sensing record all resulted from different amounts and sequences of rainfall so there is a need to standardize the vegetation cover change data if they are to be compared. Thirdly, there are few high-quality rainfall recording stations in central Australia so the precise amount of rainfall producing a given pulse is often unknown. Also, even where rainfall is known at a specific location, it cannot be extrapolated with sufficient accuracy to describe the actual variation across the 60 000 km2 region. This prevents the use of rainfall in the standardization procedure. We dealt with these difficulties as follows.

On average, the increase in plant cover during a growth pulse varies inversely with initial cover in a more or less linear fashion (Fig. 2) (see Pickup, Bastin & Chewings 1994 for an explanation of this pattern). The slope of the relationship seems to increase with the amount of rainfall and there may also be small differences in the position of the intercept. One way of standardizing the vegetation response for different rainfalls might be to derive a relationship between rainfall magnitude and the slope of the vegetation response curve. However, where rainfall is unknown or uncertain, this is not feasible.


Figure 2. Increases in vegetation cover in relation to initial plant cover, as measured by PD54 cover values, for major rainfalls in March 1983 (▪), September 1986 (+), June 1988 (;) and March 1989 (E) for an area with sandy soils in central Australia occupied by a mix of ephemeral and perennial grasses and forbs. Data were derived from a time series of Landsat MSS data and are expressed in units of the PD54 vegetation index with values of 80 = no cover and 254 = 100% cover.

Download figure to PowerPoint

An alternative approach is to examine how the vegetation response to rainfall changes over time across a grazing gradient. For example, as a landscape becomes degraded, after allowing for variations in initial cover, we would expect the greatest reduction in response to rainfall in those areas close to water since that is where grazing impact is highest. Alternatively, where a landscape has some capacity to recover from grazing, the grazing gradient should tend to disappear after rainfall, which requires that the vegetation response is greater close to water than at a distance. These patterns can be detected without information on rainfall by calculating the ratio of the vegetation response in the high grazing impact area close to water to that in a neighbouring benchmark region where the impact of grazing is limited or significantly smaller. Where the ratio decreases over time, the grazing gradient is intensifying, and the landscape is progressively degrading. Where it increases, the landscape is showing progressively more ability to respond to rainfall in the more intensively grazed areas and the grazing gradient is disappearing. If the ratio remains constant, the landscape is neither degrading nor improving.

Some flexibility is possible in defining these areas. For example, sheep do not graze as far from water as cattle and some cattle breeds will walk further than others. For conditions prevailing in central Australia where most grazing activity occurs within 4–6 km from water, a good approximation is to use the area within 4 km of water as the high grazing impact area and select the area beyond 6 or 8 km as the benchmark. However, there are a few extreme situations where both areas need to be closer in or further out from watering points.

The method is suitable for use with the full range of grazing gradient types recognized by Pickup, Bastin & Chewings (1994) because, as airborne video data show, most of the response to rainfall over the growth periods used consists of an increase in herbage cover rather than the greening up of trees and shrubs (G. Pickup, G. N. Bastin & V. H. Chewings, unpublished data). Also, the PD54 index is useful in this regard because it was developed to detect changes in vegetation cover and to reduce the influence of simultaneous shifts in greenness on the cover signal. Concentrating on the herbage response means that the effect of degradation will be detected even with inverse grazing gradients where total cover decreases with distance from water due to an increase of woody shrubs in the vicinity of water. These shrubs suppress herbage growth, creating a reduced total vegetation response to rainfall close to water.

Implementing the method is relatively simple and uses the standard grazing gradient software described by Bastin, Chewings & Pearce (1996). Using the GIS, the vegetation response is calculated by subtracting a vegetation index map showing cover before a growth pulse from a similar map containing data on cover after the pulse. Masks are applied stratifying the response data by landscape type to remove the effects of differences in vegetation cover and soil colour and by distance from water. A vegetation response data set is then derived for the whole of each stratified area by calculating the mean vegetation response for each one of a set of initial cover classes and a weighted vegetation response ratio for each growth pulse calculated as:

  • image

in which R is the ratio; Rg and Rb are the mean vegetation response values for each initial cover class, i, for the high grazing impact and benchmark areas, respectively; Ag is the area occupied by each initial cover class; and n is the number of classes. Thus, the stratified areas remain the same over time and the whole landscape within those areas is analysed. This removes much of the localized variability associated with analysis of a limited number of individual and supposedly representative sites common in rangeland vegetation analysis (e.g. Bastin et al. 1993).

The equation allows for the effect of variations in tree and initial herbage cover on the growth pulse by calculating a ratio for each of a series of initial cover classes, weighting that ratio by area and then summing values over the range of cover classes. It circumvents the need for rainfall data by comparing data for particular water points with that of surrounding areas where rainfall should be similar. There is no need to convert vegetation index values to percentage cover or biomass because the ratio is dimensionless. The use of a ratio also reduces or removes any errors or inconsistencies caused by differences in vegetation index values calculated from remotely sensed data acquired at different times. These might arise because of problems with atmospheric correction, uncorrected sensor drift, differences in scene brightness not removed by sun angle corrections, and variations in the amount of shadow present that cannot be removed by sun angle correction, etc. We expect these problems to be minor given the rigorous standardization procedures applied to the remotely sensed data (Pickup, Chewings & Nelson 1993). However, ratioing makes for a fairly robust procedure even where such standards have not been applied.

A further advantage of ratioing is that, because it compares like with like, there is no problem with differences between vegetation response to summer and winter rainfall. These are not of great significance in the rangelands of central Australia, where lower temperatures during winter months result in germination and growth of different species rather than setting limits to increase in cover. However, they could be of importance in southern Australia where winters are cooler.

Application to specific cases

  1. Top of page
  2. Abstract
  3. Introduction
  4. Development of methods
  5. Application to specific cases
  6. Conclusions
  7. References

To illustrate the trend detection method, we have applied it to a number of large paddocks or grazing units in central Australia where a particular type of change can be inferred from historical information or is strongly suspected because of a change in the management approach. No other testing method is feasible because the ground-based condition monitoring network in the region has proved unable to provide definitive information on trends (Bastin 1992). Indeed, suitable ground data are unavailable more generally (Jessup, Andrew & Lay 1994). The examples have been selected to show the range of behaviours that might be expected during analysis of a whole region and the problems of interpretation that might be encountered. We also show how the method may be varied when paddocks are not sufficiently large to define a benchmark area far from watering points. In most cases, we have concentrated on landscape types most favoured for grazing since it is here that degradation and recovery are likely to be most pronounced.

Case 1

Case 1 is a 460-km2 paddock on the south-eastern edge of the MacDonnell Ranges about 100 km east of Alice Springs. The paddock is adjacent to the Todd River and has alluvial landscapes, extensive foothill alluvial fan systems and low rises. The alluvial landscapes are highly favoured by cattle and have clay or clay loam soils supporting perennial and ephemeral grasses, or sandy soils occupied by open woodland with an understorey of perennial grasses, ephemeral herbs and grasses, and woody shrubs.

The paddock has three permanent water points, all of which were developed during the 1960s drought, and the paddock was stocked at that time. At least one water point was shut down during the 1970s and, for part of that time, there were no cattle in much of the analysis area. The property changed hands in 1979, the closed-down water point was reactivated and relatively high stocking rates were maintained until the drier years of the 1990s.

The result of this series of events has been the development of a normal grazing gradient on the alluvial landscapes that is maintained even after significant rainfall. The gradient over the first 5 km from water seemed to intensify after 1983, although it produced a relatively flat curve after the 1995 rainfall pulse (Fig. 3a). The presence of this gradient was verified in a major ground sampling exercise by Bastin, Sparrow & Pearce (1993).


Figure 3. (a) Gradients in vegetation cover in relation to distance from water in Case 1 after five major growth pulses in March 1983 (▪), September 1986 (+), June 1988 (;), March 1989 (E) and May 1995 (R). Vegetation cover data are expressed in PD54 units. (b) Vegetation response as measured by R (see text for details) from 1983 to 1995 using the area 0–4 km from water as the high grazing impact zone and the areas beyond 4 km (▪), 6 km (+) and 8 km (E) as the low grazing impact benchmark zone.

Download figure to PowerPoint

The associated graph of R-values compares growth pulse data in the first 4 km from water with three different benchmark areas, starting at 4 km, 6 km and 8 km from water, respectively (Fig. 3b). Irrespective of which benchmark area was used, the R-values were less than 1 except in a single case. This indicates a severe grazing gradient maintained over a long period where the vegetation response is less closer to water than further out. There was also a downward trend until the 1995 growth pulse, showing that grazing impact increased through the 1980s.

The 8-km R-values lie below those derived using the 6-km benchmark which, in turn, are less than the 4-km benchmark values. This occurs because grazing partly extends into the benchmark area, so the contrast between recovery in the first 4 km and the benchmark areas will be progressively greater for those benchmarks that start further out from water. Even so, the pattern of change over time is similar in all three cases, suggesting that there can be some flexibility in the selection of benchmark areas.

The 1995 growth pulse produced a flatter post-rainfall cover gradient than the previous three events (Fig. 3a). This may mean that there was some recovery from grazing through an increase in the capacity of vegetation to respond to rainfall close to water. Even so, the R-values are all less than 1 so we are dealing with a slowing down of the rate of degradation rather than a restoration of land condition. Alternatively, it might be that the area of maximum grazing impact moved further out from water and affected the area we have used as a benchmark. A further explanation might be that the rainfall which produced the growth pulse was too small to generate sufficient recovery from grazing to be used in this type of analysis. However, the Monthly Weather Review for January 1995 (Bureau of Meteorology 1995) suggests that the area had between 100 and 200 mm of rainfall, which should be sufficient for a good response.

Comparison of the pre- and post-growth cover values in the 1989 and 1995 pulses provides some support for the idea of a slight recovery implied by the increase in the 1995 values (Fig. 4). The 1989 curves are basically parallel, indicating similar levels of response to rainfall over most of the first 10 km from water. The curves for the 1995 pulse show greater vegetation growth close to water than further out, which is characteristic of recovery even though the amount of growth was very much smaller than in 1989. At the same time, the convergence of R-values in 1995 is consistent with either a shift of the area of maximum grazing impact into the benchmark area, or very limited growth in response to a small rainfall that did not allow sufficient recovery to remove temporary grazing effects. Our conclusion for this landscape type in this paddock is that there was a downward trend in condition between 1983 and 1989 which may have slowed in the 1990s (see below).


Figure 4. Vegetation cover in relation to distance from water for Case 1 before and after the growth pulses in 1989 and 1995. Symbols indicate March 1989 (▪), May 1989 (+), December 1994 (;) and May 1995 (E). Vegetation cover data are expressed in PD54 units.

Download figure to PowerPoint

Case 2

Most of the Todd River floodplain has been grazed for a longer period than the paddock examined in Case 1. It was therefore interesting to see how similar areas with a significantly greater grazing impact behaved through the same rainfall sequence.

Case 2 is a 390-km2 paddock on the Todd River floodplain closer to Alice Springs but with a very similar mix of landscape types to Case 1. The paddock has five permanent water points based on bores and several dams. While two of these were established in the late 1970s, the rest were in place by the early 1960s and at least one is pre-World War II. Anecdotal evidence (e.g. Hayes 1987) indicates that this paddock has been grazed continuously (sometimes quite heavily) for a much longer time than the paddock in Case 1. Initially, this would have been from a few water points but by the late 1970s all parts of the paddock were within range of water.

Post-rainfall grazing gradients for the same landscape types as those used in Case 1 are shown in Fig. 5(a). A strong gradient, extending to 12 km from the water point, was maintained after every growth pulse. This indicates a substantial grazing impact and little recovery in spite of the two large rainfall events. R-values, comparing the 0–4-km zone with benchmark areas beyond 4 km and 6 km, confirm this (Fig. 5b). All are less than 1 and there is virtually no change over time in the 4-km and 6-km benchmark data compared with the results for Case 1. There is also no sign of recovery in the 1995 event, which suggests that the improvement shown for that event in Case 1 is genuine.


Figure 5. (a) Gradients in vegetation cover in relation to distance from water in Case 2 after five major growth pulses in March 1983 (▪), September 1986 (+), June 1988 (;), March 1989 (E) and May 1995 (R). Vegetation cover data are expressed in PD54 units. (b) Vegetation response as measured by R (see text for details) from 1983 to 1995 using the area 0–4 km from water as the high grazing impact zone and the areas beyond 4 km (▪) and 6 km (+) as the low grazing impact benchmark zone.

Download figure to PowerPoint

Case 3

Case 3 is a 336-km2 paddock about 250 km south of Alice Springs. It occupies an area of lightly dissected tablelands with calcareous soils. There are also areas of sand sheet, remnant rocky tablelands and alluvial areas associated with creek systems. We concentrated on the calcareous soils and the sand sheets.

The calcareous areas have either loamy or clay soils. Loamy soils support scattered trees and woody shrubs or moderately dense chenopod shrubs, particularly bluebush Maireana astrotricha (Dunlop et al. 1992), with mixed herbage and grass species also present. The clay soils carry mainly bluebush and other chenopod species. Both soil types are highly attractive to cattle, while the loamy soils are also infested with feral rabbits. Both units are fragile and have been extensively degraded but, even so, the sparse ephemeral herbage remains very attractive for grazing. The sand sheets have more trees and shrubs, more cover generally, and more perennial grasses in the herbage layer but are less attractive to cattle than the calcareous areas. These areas are also considered to be tolerant of grazing.

The area had one permanent and two semi-permanent water points until 1983. At that time, the present paddock was fenced and since then, additional water points have been added by reticulation through pipelines. This has dispersed grazing pressure considerably. The property was engaged in the eradication of bovine brucellosis until about 1987 or 1988 so quite high numbers of cattle had to be held in the paddock. Since that time, cattle numbers have been reduced. Rabbits have been progressively controlled in the northern and eastern parts of the paddock by warren ripping. Some attempts at land reclamation have been made but with limited success until recent times. There have been very few significant rainfall events since 1989 so, if vegetation cover has improved, then it is likely to be mainly due to fewer cattle and the control of rabbits.

The two calcareous landscapes and the sand sheets respond to, and recover from, grazing in different ways and have therefore been examined separately. On the calcareous soils, a normal grazing gradient has developed that has been maintained even after some very substantial rainfalls (Fig. 6a). For example, the 1989 growth pulse is thought to be among the best in the last 30–40 years by people with extensive local knowledge. The sand sheets show no major grazing impact after large growth pulses (Fig. 6b). However, there is some evidence of an inverse gradient in which cover decreases rather than increases with distance from water. This is often an indication of a grazing-induced build-up of unpalatable woody shrubs that can reduce herbage growth (e.g. Pickup, Bastin & Chewings 1994; Booth, Sänchez-Bayo & King 1996) and is typical of sandy soils throughout the arid and semi-arid areas of Australia.


Figure 6. (a) Gradients in vegetation cover in relation to distance from water in Case 3 on the calcareous soils after two major and three lesser growth pulses. The major pulses were recorded in March 1983 (▪) and March 1989 (E). Smaller pulses were measured in September 1986 (+), June 1988 (;) and May 1995 (R). (b) The grazing gradients on the sandy soils at the same time. Vegetation cover data are expressed in PD54 units. (c) Vegetation response as measured by R (see text for details) from 1983 to 1995 for the calcareous soils (▪), and the sandy soils (+) using the area 0–4 km from water as the high grazing impact zone and the area beyond 8 km as the low grazing impact benchmark zone.

Download figure to PowerPoint

The R-values for the calcareous landscapes are less than 1 and decrease with time until the 1995 growth pulse (Fig. 6c). This is consistent with a strong and intensifying grazing gradient. The 1995 pulse produced a small reversal in this decline in condition. However, it is difficult to say whether this reflects the changes in grazing management described above as it is subject to the same uncertainty as Case 1. The sand sheets showed a similar decline in condition in the early 1980s but the R-values lie above, or close to, 1 suggesting a greater level of resilience than on the calcareous areas (Fig. 6c). The improvement in condition also began at an earlier stage on the sand sheets and by 1995 was very substantial, even with a relatively small rainfall event. We attribute this change to the different grazing management that prevailed after 1987 together with the greater inherent capacity of this landscape to recover from grazing. However, it would be useful to include a future, larger growth pulse in the analysis for confirmation.

Case 4

All of the previous cases describe areas that are moderately to highly attractive to cattle so grazing impact is potentially high. In Case 4, we describe the behaviour of a landscape that only gets limited usage because its forage is relatively unpalatable.

The area in question is part of a 155-km2 paddock about 70 km north-west of Alice Springs on the northern edge of the Chewings Ranges. The paddock has five water points, two of which are pre-World War II, two are from the 1950s and one from the late 1960s. The landscape has a mix of ancient fan surfaces with heavily weathered red earths together with more recent alluvial and flood-out deposits (described by Low 1978; Pickup & Chewings 1988). The red earths, which we examine in this section, make up about half of the paddock, are fairly resistant to erosion, and are occupied by mulga Acacia aneura (Dunlop et al. 1992) woodland with an understorey of perennial grasses and ephemeral herbs and grasses. This vegetation type is unattractive to cattle, which tend to use it mostly during dry times when forage is depleted on other more preferred grazing areas (Low, Dudzinski & Muller 1981).

Limited grazing use and the erosion-resistant soils suggest that the effects of grazing on vegetation cover will be small and the area should recover well after rainfall. The grazing gradient curves confirm this by showing virtually no change in cover with distance from water except in the first kilometre after each of the five growth pulses examined (Fig. 7a). The R-curves, which are based on a 6-km benchmark area, also show that the landscape is resilient at the level of grazing experienced, all values being slightly greater than 1 (Fig. 7b). This level of resilience is sufficient to allow removal of the temporary grazing gradients that develop between growth pulses. There is also no trend in the R-values, which indicates that the area is stable, as might be expected.


Figure 7. (a) Gradients in vegetation cover in relation to distance from water in Case 4 after five major growth pulses in March 1983 (▪), September 1986 (+), June 1988 (;), March 1989 (E) and May 1995 (R). Vegetation cover data are expressed in PD54 units. (b) Vegetation response as measured by R (see text for details) from 1983 to 1995 using the area 0–4 km from water as the high grazing impact zone and the area beyond 6 km as the low grazing impact benchmark zone.

Download figure to PowerPoint

Case 5

In all previous cases, water point layout and paddock size were such that there was an area sufficiently distant from water to provide a benchmark area that experienced limited grazing. In Case 5, we examined a landscape type attractive for grazing but where the distribution of water points means that there is no part of this landscape further than 6 km from water. The benchmark zone therefore has to be set at 4 km from water (or closer), and even then the available area is small. This means that local variability within the landscape type can affect the results and may produce erratic patterns in R-values. Also, R-values will be closer to 1 than if a benchmark area more distant from water is used.

The area examined in this case study was the sandy alluvial landscape of the same paddock used in Case 4. It is an open woodland of mainly Acacia species with an understorey of perennial and ephemeral grasses and ephemeral herbs generally attractive to cattle (Low, Dudzinski & Muller 1981). The area has proved resilient to grazing (Fig. 8a) but, more recently, a grazing gradient has developed and been maintained even after some relatively large growth pulses.


Figure 8. (a) Gradients in vegetation cover in relation to distance from water in Case 5 after five major growth pulses in March 1983 (▪), September 1986 (+), June 1988 (;), March 1989 (E) and May 1995 (R). Vegetation cover data are expressed in PD54 units. (b) Vegetation response as measured by R (see text for details) from 1983 to 1995 using the area 0–3 km from water as the high grazing impact zone and the areas beyond 3 km (▪) and 4 km (+) as the low grazing impact benchmark zone. (c) Increases in vegetation cover for a high grazing impact area set at 0–4 km from water in 1983 (▪) and 1989 (+) and for a benchmark area beyond the 4 km point in 1983 (;) and 1989 (E). (d) The same data but with the boundary between the high grazing impact and benchmark area set at 3 km from water.

Download figure to PowerPoint

Two sets of R data are presented in Fig. 8(b) to illustrate both the change in state and the problems of variability created by having access to only a small benchmark area. The first was derived by using the 0–3 km zone as the area of high grazing impact and the zone beyond 3 km as the benchmark. The second used the 4-km point as the boundary between the high grazing impact zone and the benchmark.

Both sets of R data produce a similar pattern, although the 3-km values are lower, as would be expected from using areas closer to water. Values for the growth pulses in 1983, 1986 and 1988 are substantially greater than 1, even though the benchmark areas are relatively close to water, showing that the landscape has considerable potential to recover from grazing. After the 1988 rainfall event, there was a substantial drop in the values followed by a partial recovery. This coincided with a change in management personnel on the pastoral lease but there are insufficient data to confirm whether the drop was a temporary aberration or marked the beginning of a loss in land condition. At least one more major growth pulse should be examined before drawing a firm conclusion.

The effect of using small areas on the Rg and Rb values can be seen by comparing the vegetation response for the two pairs of high grazing impact and benchmark areas (Fig. 8c,d). Bastin et al. (1993) have previously examined the impact of local variability in grazing gradient analysis and shown that it is desirable to have areas of 7–15 km2 or more to minimize it. The individual cover classes in the benchmark area of the 4-km data set occupy areas of 0·1–1 km2, hence the very different and highly variable vegetation response when compared with that present in the area of high grazing impact. The 3-km data set has areas of 1–17 km2 in the individual cover classes of the benchmark area. This is still small in some cases but shows a pattern of vegetation response closer to the normal inverse linear relationship depicted in Fig. 2.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Development of methods
  5. Application to specific cases
  6. Conclusions
  7. References

The examples presented above show that analysis of how vegetation growth pulses vary along grazing gradients can detect change in rangeland condition. The procedure is relatively simple and, in Australia at least, can be implemented using archived data from remote sensing satellites. Information on rangeland trends attributable to grazing is therefore immediately available, whereas monitoring by other methods may not produce results for years while necessary data are accumulating.

The sensitivity of the method seems high in comparison with other techniques, particularly where the landscape contains a mix of pasture types and natural soil erosion and deposition features. For example, in the paddock described in Cases 4 and 5, a ground-based monitoring network operated over the period 1982–89 in which vegetation species data were collected from 150 locations on a grid covering five small to major growth pulses. Even with such a large number of sites covering a wide range of distances from water, it did not prove possible to identify a short-term grazing effect, let alone a trend in landscape condition (Friedel, Pickup & Nelson 1993). This problem arose because the landscape is highly complex and grazing effects were confounded by landscape variability, even though the spatial and temporal sampling frequency far exceeded that which might be used in an operational network.

Our method is efficient when compared with conventional ground-based vegetation sampling. For example, in Case 1, measurement of the grazing gradient on a single occasion (Bastin, Sparrow & Pearce 1993) required about 7 man-days even though the area lacks the complexity of the landscapes examined in most other cases in the paper. Repeating this exercise for more than one growth pulse and for the many locations required in an operational network is logistically impossible. By comparison, once the GIS is set up, which costs about the same as installing a ground-based network (Bastin et al. 1993), a grazing gradient analysis of trend can be done for many locations in a few hours.

The trend detection method has wide applicability in Australian rangelands. Other variants of the grazing gradient method have been used successfully in regions with median annual rainfalls varying between 150 and 450 mm and extending over latitudes from 18°S to 23°S. We would expect the trend detection method to work in these situations, although in monsoonal areas it might be more appropriate to measure gradients at the end of the wet season. We would not expect it to work in its current form in the higher rainfall regions of northern Australia where grazing is much less dependent on a small number of artificial water points. The other main restriction on use is paddock size, since paddocks must be large enough to have a suitable benchmark region present. Given that sheep range over shorter distances than cattle, the method can be applied to smaller paddocks in sheep-grazing areas than in cattle-producing regions. However, paddocks are probably still too small in the subhumid rangelands of eastern Australia. This suggests that the method could be suitable for perhaps half to two-thirds of rangeland Australia, although further testing is recommended.

The method also has potential for application outside Australia. Arid and semi-arid rangelands in parts of North America, South America and Africa are managed on a large enough scale to allow development of the spatial patterns of grazing impact that the method exploits. This can occur under both commercial and subsistence grazing. The method might also be used to monitor patterns of human impact around villages in traditional societies where rainfall data for trend analysis might be difficult to obtain.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Development of methods
  5. Application to specific cases
  6. Conclusions
  7. References
  • ANZECC/ARMCANZ, Joint Working Group 1996Draft National Strategy for Rangeland Management . Department of Environment, Sport & Territories, Canberra, Australia.
  • Bastin, G.N. 1989Centralian Range Assessment Program . Northern Territory Department of Primary Industry and Fisheries Technical Bulletin 151. Northern Territory Government, Alice Springs, NT, Australia.
  • Bastin, G.N. 1992 The Centralian range assessment program. Proceedings of the 5th Australian Soil Conservation Conference, Vol. 7, Range Monitoring Workshop(eds G.J.Hamilton, K.M.Howes & R.Attwater), pp. 3539. Department of Agriculture, Perth, Australia.
  • Bastin, G.N., Chewings, V.H. & Pearce, G. 1996A Manual for the Landscape-Scale Assessment of Cattle Grazing Impact in Northern South Australia Using Satellite Data and Grazing Gradient Methods . CSIRO Division of Wildlife and Ecology, Centre for Arid Zone Research. Alice Springs, NT, Australia.
  • Bastin, G.N., Pickup, G., Chewings, V.H. & Pearce, G. 1993 Land degradation assessment in central Australia using a grazing gradient method. The Rangeland Journal, 15, 190216.
  • Bastin, G.N., Sparrow, A.D. & Pearce, G. 1993 Grazing gradients in central Australian Rangelands: ground verification of remote sensing based approaches. The Rangeland Journal, 15, 217233.
  • Booth, C.A., Sänchez-Bayo, F. & King, G.W. 1996 Establishment of woody weeds in western New South Wales. II. Growth and competitive potential. The Rangeland Journal, 18, 8098.
  • Bureau of Meteorology 1995 Monthly Weather Review, Northern Territory January 1995. Bureau of Meteorology, Darwin, NT Australia.
  • Cardy, F. 1994 Desertification. Our Planet, 6, 4.
  • Condon, R.W. 1986 Recovery of catastrophic erosion in western New South Wales. Proceedings of 2nd International Rangeland Congress. Rangelands: a Resource Under Siege. (eds P.J.Joss, P.W.Lynch & O.B.Williams), pp. 39. Australian Academy of Science, Canberra, Australia.
  • Cridland, S. & Stafford Smith, D.M. 1993Development and Dissemination of Design Methods for Rangeland Paddocks Which Maximise Animal Production and Minimise Land Degradation. Western Australian Department of Agriculture Miscellaneous Publication 42/93. Government of Western Australia, Perth, WA, Australia.
  • Cunningham, G.M. 1996 Centralia revisited: resurvey of rangeland sites in the Alice Springs area. 1993–95. Range Management Newsletter, 96, 46.
  • DeAngelis, D.L. & Waterhouse, J.C. 1987 Equilibrium and non-equilibrium concepts in ecological models. Ecological Monographs, 57, 121.
  • Department of Environment, Sport, & Territories 1996Australia: State of the Environment. CSIRO Publishing, Melbourne, Australia.
  • Duckett, N.J., Holm, A. & Mc R. 1996 Is the vegetation of the Western Australian southern shrublands changing over time? Proceedings of 9th Australian Rangeland Society Biennial Conference. Focus on the Future – the Heat Is on! (eds P.Hunt & R.Sinclair), pp. 139140. Australian Rangeland Society, Port Augusta, SA, Australia.
  • Dunlop, C.R., Leach, G.J., Latz, P.K. & Barritt, M.J. 1992 Checklist of Vascular Plants of the Northern Territory, Australia. Conservation Commission of the Northern Territory.
  • Eldridge, D.J. & Rothon, J. 1992 Runoff and sediment yield from a semi-arid woodland in eastern Australia. I. The effect of pasture type. The Rangeland Journal, 14, 2639.
  • Ellis, J.E., Coughenour, M.B. & Swift, D.M. 1993 Climatic variability, ecosystem stability, and the implications for range and livestock development. Range Ecology at Disequilibrium (eds R.H.Behnke, I.Scoones & C.Kervin), pp. 3141. Overseas Development Institute, London, UK.
  • Foran, B.D., Bastin, G.N., Remenga, E. & Hyde, K.W. 1982 The response to season, exclosure and distance from water of three central Australian pasture types grazed by cattle. Australian Rangeland Journal, 4, 515.
  • Foran, B.D., Bastin, G.N. & Shaw, K. 1986 Range assessment and monitoring in arid lands: the use of classification and ordination in range survey. Journal of Environmental Management, 22, 6784.
  • Frank, T.D. 1984 The effect of change in vegetation cover and erosion patterns on albedo and texture of Landsat images in a semi-arid environment. Annals of the Association of American Geographers, 74, 393407.
  • Friedel, M.H. 1984 Biomass and nutrient changes in the herbaceous layer of two central Australian mulga shrublands after unusually high rainfall. Australian Journal of Ecology, 9, 2738.
  • Friedel, M.H. 1985 The population structure and density of central Australian trees and shrubs, and relationships to range condition, rabbit abundance and soil. Australian Rangeland Journal, 7, 130139.
  • Friedel, M.H., Pickup, G. & Nelson, D.J. 1993 The interpretation of vegetation change in a spatially and temporally diverse arid Australian landscape. Journal of Arid Environments, 24, 241260.
  • Graetz, R.D., Fisher, R.P. & Wilson, M.A. 1992Looking Back: the Changing Face of the Australian Continent, 1972–92. CSIRO Office of Space Science and Applications, Canberra, Australia.
  • Graetz, R.D., Pech, R.P. & Davis, A.W. 1988 The assessment and monitoring of sparsely vegetated rangelands using calibrated Landsat data. International Journal of Remote Sensing, 9, 12011222.
  • Graham, O.P., Emery, K.A., Abraham, N.A., Johnston, D., Pattemore, V.J. & Cunningham, G.M. 1990Land Degradation Survey New South Wales 1987–88. Soil Conservation Service of New South Wales, Sydney, Australia.
  • Griffin, G.F. & Friedel, M.H. 1985 Discontinuous change in central Australia: some implications of major ecological events for land management. Journal of Arid Environments, 9, 6380.
  • Hacker, R.B., Wang, K.M., Richmond, G.S. & Lindner, R.K. 1991 IMAGES: an integrated model of an arid grazing ecological system. Agricultural Systems, 37, 119163.
  • Hayes, E. 1987 The changes I have observed in the rangeland areas of the MacDonnell Ranges. 1927–87. Range Management Newsletter, 87 (3), 67.
  • Hellden, U. 1991 Desertification – time for an assessment? Ambio, 20, 372383.
  • Hobbs, T.J., Sparrow, A.D. & Landsberg, J.J. 1994 A model of soil moisture balance and forage plant growth in the arid rangelands of central Australia. Journal of Arid Environments, 28, 281298.
  • Holling, C.S. 1973 Resilience and stability of ecological systems. Annual Review of Ecology and Systematics, 4, 123.
  • Holm, A., Mc R., Burnside, D.G. & Mitchell, A.A. 1987 The development of a system for monitoring trends in the arid shrublands of Western Australia. Australian Rangeland Journal, 9, 1420.
  • Jessup, P.J., Andrew, M.H. & Lay, B. 1994 Indicators for range assessment. Clean Product, Clean Country, Clear Profit, Working Papers of the 8th Australian Rangeland Society Biennial Conference (ed. G.N.Bastin), pp. 5153. Australian Rangeland Society, Katherine, NT, Australia.
  • Lamprey, H.F. 1988 Report on the desert encroachment reconnaissance in northern Sudan: 21 October to 10 November 1975. Desertification Control Bulletin, 17, 17.
  • Low, W.A. 1978The Physical and Biological Features of Kunoth Paddock in Central Australia . CSIRO (Australia) Division of Land Resources Management Technical Paper 4. CSIRO, Canberra, Australia.
  • Low, W.A., Dudzinski, M.L. & Muller, W.J. 1981 The influence of forage and climatic conditions on range community preference of Shorthorn cattle in arid central Australia. Journal of Applied Ecology, 18, 1126.
  • Palmer, D. 1991 Western New South Wales – a miracle of recovery. Australian Journal of Soil and Water Conservation, 4, 48.
  • Perkins, J.S. & Thomas, D.S.G. 1993 Spreading deserts or spatially confined environmental impacts? Land degradation and cattle ranching in the Kalahari Desert of Botswana. Land Degradation and Rehabilitation, 4, 179194.
  • Pickard, J. 1993 Western New South Wales – increased rainfall, not a miracle, leads to recovery. Australian Journal of Soil and Water Conservation, 6, 49.
  • Pickup, G. 1989 New land degradation survey techniques for arid Australia – problems and prospects. Australian Rangeland Journal, 11, 7482.
  • Pickup, G. 1994 Modelling patterns of defoliation by grazing animals in rangelands. Journal of Applied Ecology, 31, 231246.
  • Pickup, G. 1995 A simple model for predicting herbage production from rainfall in rangelands and its calibration using remotely-sensed data. Journal of Arid Environments, 30, 227245.
  • Pickup, G. 1996 Estimating the effects of land degradation and rainfall variation on productivity in rangelands, an approach using remote sensing and models of grazing and herbage dynamics. Journal of Applied Ecology, 33, 819832.
  • Pickup, G., Bastin, G.N. & Chewings, V.H. 1994 Remote sensing-based condition assessment for non-equilibrium rangelands under large-scale commercial grazing. Ecological Applications, 4, 497517.
  • Pickup, G. & Chewings, V.H. 1988 Estimating the distribution of grazing and patterns of cattle movement in a large arid zone paddock: an approach using animal distribution models and Landsat imagery. International Journal of Remote Sensing, 9, 14691490.
  • Pickup, G., Chewings, V.H. & Nelson, D.J. 1993 Estimating changes in vegetation cover over time in arid areas from remotely sensed data. Remote Sensing of Environment, 43, 243263.
  • Purvis, J.R. 1986 Nurture the land: my philosophies of pastoral management in central Australia. Australian Rangeland Journal, 8, 110117.
  • Senft, R.L., Coughenour, M.B., Bailey, D.W., Rittenhouse, L.R., Sala, O.E. & Swift, D.M. 1987 Large herbivore foraging and ecological hierarchies. Bioscience, 37, 789799.
  • Stafford Smith, M. & Pickup, G. 1990 Pattern and production in arid lands. Proceedings of the Ecological Society of Australia, 16, 195200.
  • Tothill, J.C. & Gillies, C. 1992 The Pasture Lands of Northern Australia, their Condition, Productivity and Sustainability. Tropical Grasslands Society of Australia Occasional Publication 5. Cranbrook Press, Toowoomba, Queensland, Australia.
  • Warren, P.L. & Hutchinson, C.F. 1984 Indicators of rangeland change and their potential for remote sensing. Journal of Arid Environments, 7, 104126.

Received 7 June 1997; revision received 12 January 1998