Previous PageTable Of ContentsNext Page


P.W.G. Sale, M.R. McCaskill and G.J. Blair

Department of Agronomy and Soil Science, University of New England. Armidale. 2351

Reports of disappointing performances of improved pasture swards are now being regularly received from landholders and agronomists in the New England. .In particular, the persistence and productivity of our major pasture legume, white clover (Trifolium repens), appears to have declined.

A number of potential agents have been put forward to explain why this perennial pasture legume is less persistent than it was in the early years of pasture improvement (1950s and early 1960s), when clover would grow ‘up to your knees’ when the country was topdressed with superphosphate and seed. The list of possible causal agents includes virus attack, nematode infestations in light country, reduced vigour due to outcrossing with less productive native clovers or to reduced fertiliser usage. However, one factor that is implicated most frequently is the effect of climate, in that the recent run of dry years in the 1980s has severely restricted white clover persistence. This paper outlines the results of recent research at the University of New England (McCaskill, 1987) which postulated that medium-term changes in climate have occurred, and that such changes would have a marked impact on white clover persistence.

Evaluation of medium-term climatic changes in the New England

A number of studies have been carried out which indicate that changes in rainfall patterns in eastern Australia have occurred this century. The average rainfall in the 30-year period between 1881 and 1910 was higher than that for the following 30-year period (Ralph, 1983). Another study revealed that a swing back towards the more favourable climate of the late 19th Century took place in parts of eastern Australia from the late 1940s to the early 1970s (Pittock and Salinger, 1982), a finding that was also substantiated by Russell (1981).

A number of questions arise from such studies. If changes have taken place in rainfall patterns, what then constitutes the “normal season”? If the seasons of the past were now repeated, how then would white clover produce and persist? This would depend on both the intensity and duration of moisture stress encountered by the white clover plant had it been growing under that climatic regime. These early studies on changing rainfall patterns cannot assess the impact of past rainfall on white clover performance, as they only utilise annual or summer rainfall totals.

A novel approach to assessing past rainfall patterns on white clover performance was used by McCaskill (1987). By using the water balance routine in a mechanistic model which stimulates the growth of temperate pastures, McCaskill was able to predict the level of available moisture in the soil profile for a given combination of rainfall, temperature and evaporation data. The infiltration, surface runoff, lateral and deep leaching components of a given rainfall event were simulated on a daily time-step. Similarly, the loss of moisture from the soil by surface evaporation and transpiration from different soil depths was also calculated. The model was then used to calculate what is termed the number of critical clover days per year. A critical clover day (CCD) is a day between October and May when the soil moisture content falls below 17% of its available waterholding capacity. This would impose a severe moisture stress on the white clover plant during that time of the year when its growth is not limited by low temperature. The greater the number of CCDs in a year, the greater the duration of moisture stress; a series of years with many CCDs would indicate a time period when white clover persistence would decline due to clover death and the depletion of seed reserves.

Daily rainfall data back to 1858 for Armidale, NSW, were available although some years were missing during the 1860s and 1870s. Each year was simulated by the model and the average number of critical clover days per year was calculated.

The analysis shows that over the period from 1858 to 1986 there was an average of 19.4 CCDs per year (Table 1). In the period 1948-1964, when the major expansion of pasture improvement took place in the New England, there was an average of 2.1 CCDs per year. Similarly, in the period 1950-1979 when most temperate pasture species were selected, the figure was 6.3 CCDs per year.

Table 1. Mean rainfall and number of critical clover days (CCD) per year for each decade at Armidale, NSW.


Rainfall (mmyr )


1850 - 1860s







































Mean - overall



- 1948-64



- 1950-79



Implications of medium-term climatic change in the New England

It would appear that the rainfall patterns during the 1950s, 1960s (apart from 1965) and the 1970s were extremely favourable for the production and persistence of white clover in the New England. Now, in the 1980s, the rainfall distribution is returning to that which approximates the longer-term pattern (9 decades out of 12). This is believed to be the major reason why white clover has essentially disappeared from our pastures in recent years. The same can be said for our ryegrasses that were selected in the 1950s and deemed to be suitable for the New England, but have disappeared in the 1980s.

It is imperative that funding support be given to research projects that evaluate alternative species and cultivars for their ability to perform under the drier climatic regimes that we are now experiencing in the north. Obviously, with such research, a balance must be struck between productivity and persistence.

Early years of pasture improvement in the New land - a favourable combination of circumstance

It is quite amazing to note how all factors essential for an expansion in the area of improved pastures did, in fact, occur together in the 1950s and early 1960s in the New England. The first factor was the development of pasture improvement technology: high performing grasses and legume species had been selected for the area, machine planting and aerial topdressing procedures were developed, the need for inputs of phosphorus and sulphur to enable satisfactory legume growth on most soils was established and the myoxma virus, released in the early 1950s, led to effective control of rabbits. Economic conditions were highly favourable -the wool boom in the early 1950s, the existence of low interest rates and favourable taxation laws encouraged significant expenditure on pasture development. However, most importantly, this early era of pasture improvement received extremely favourable climatic conditions (Table 1) which enabled white clover and other introduced species to thrive. The resulting increase in quantity and quality of pasture forage led to an increase of 66% in livestock numbers and wool production by 115% (McDonald, 1968). It transformed much of the landscape and the image of the New England region of NSW.

Medium-term climatic change in southern pasture areas?

It has been noted in recent years that legumes are not persisting in the cereal-livestock areas of southern Australia (Carter et al., 1982). In these areas, typified by a more Mediterranean climate, the major pasture legumes are the annual, self-regenerating cultivars of subclover and medics. To what extent is the poor persistence of these legumes due to a change in rainfall pattern? Reasons for the decline in persistence of medics given by Carter et al. (1982) include an increase in the occurrence of insect attack (red-legged earthmite, lucerne flea, sitona weevil), less favourable fertiliser practices (less superphosphate, more nitrogen fertilisers), changing management (increased cropping, poor grazing management, increased use of herbicides, less undersowing of cereals with medics), and a general apathy towards medics. No mention is made of the possibility that drier seasons may be a major contributor. All the ‘proper’ management procedures will not help if the lack of persistence of legumes results from the poor adaptation of cultivars to existing climatic patterns. Computer simulation procedures could answer this question for the southern pasture areas, as it has done for the Northern Tablelands of NSW.

Simulation procedures could also be used to test how different species or management practices might perform, given the climatic patterns over the short- or medium-term past. All that is required is to translate rainfall and evaporation data to moisture levels in the soil profile for a given soil/plant system, and then relate the pattern of soil moisture fluctuations to the plant performance. Obviously, an accurate understanding of the physiological responses by the plant to moisture stress during its ontogenetic development would be required. Such analyses would permit the allocation of ‘‘probabilities of success’’ to a particular species or a management option, given the climatic patterns of the past. Likewise, any recurring cyclical pattern in rainfall would become apparent from this type of work.


Changes during the 1980s have occurred in rainfall patterns in the Northern Tablelands of NSW, with a return to the less favourable climatic regimes of early decades. Pasture species such as white clover and the ryegrasses, that were selected under more favourable moisture conditions, are not persisting in our pastures today. The selection and evaluation of alternative species or cultivars that will persist in drier years is now required in the north.

Analysis of past. weather data sets using computer simulation procedures may well be a valuable exercise in the southern areas of Australia. Such an analysis would determine the extent of any medium-term changes in rainfall patterns, and the degree to which such changes might be contributing to the poor persistence of species and cultivars in southern pasture areas. The likelihood of success of management options or different species could also be evaluated for past rainfall patterns, using these procedures.


1. Carter, E.D., Wolfe, E.C. and Francis, C.M.(1982). Problems of maintaining pastures in the cereal-livestock areas of southern Australia. Proc. 2nd Aust. Agronomy Cont., Wagga Wagga, pp.68-82.

2. McCaskill, M.R. (1987). Modelling 5, P and N cycling in grazed pastures. PhD thesis, University of New England.

3. McDonald, G.T. (1968). Recent pasture development on the Northern Tablelands of New South Wales. Australian_Geographer 10:382-391.

4. Pittock, A.B. and Salinger, M.J. (1982). Towards regional scenarios for a CO -warmed earth. Climatin Change 4:23-40.

5. Ralph, W. (1983). Climatic change and Australian agriculture. Rural Research 117:4-8.

6. Russell, J.S. (1981). Geographic variation in seasonal rainfall in Australia - an analysis of the 80-year period 1895-1974. J. Aust. Inst. Agric.Sci. 47:59-66.

Previous PageTop Of PageNext Page