Agriculture consists of harnessing solar energy to produce food and fibre. The sunlight and the climate determine what can be grown, and the farming practices determine the degree of success. The success depends on many factors, such as technology, farmers’ attitudes and skills, and the difference between costs and prices.

Success is not easily obtained for in any form of land use, there are factors that create varying degrees of instability and degradation. Some of these factors are climatic e.g. droughts, floods, fire; some are management practices that influence the incidence of weeds, vermin, erosion, soil fertility, pasture decline, and others are economic e.g. costs, prices, credit facilities, farm size and tenure which often force a farmer to carry out practices such as overcropping or overgrazing, which he would not normally tolerate.

This instability and degradation however can be reversed with other strategies which emphasise soil restoration. These strategies often require a change in land use or the adoption of new soil, plant or animal technology, all of which may be costly and result in a temporary loss of production and income. Some financial incentive may therefore be required to encourage farmers to change to these new strategies.

The farmer has to try and make a living by operating within the degradation and restoration cycles. A lot of knowledge about his own property is self acquired but he hopes that research and extension can help him to balance out the fluctuations.

Currently he is bewildered by the array of research activities -research into new crops and pastures, wheat quality, weeds and plant diseases, tillage systems, stubble mulching, herbicides, fertilizer needs, grazing practices etc., all of which are carried out in small plots with each project claiming a 10-20% increase in yield. Not all farmers have the problems being investigated nor can they incorporate all the research recommendations into their farming system.

The gap between science and agricultural practice seems to be widening and it is becoming harder for the farmer to know which research will give economic benefits on his own farm. Occasionally a recommendation to adopt a new research finding requires a consequential set of changes throughout the farming system. What are the guidelines that will help him select the best ideas from research?

Agricultural production is determined by a number of factors.

a. Climate

This affects plant growth as follows:-

• the amount of sunlight influences photosynthesis, and, the daylength, and in some plants, a period of cold weather (vernalisation) determinesthe number of flower buds.
• the temperature pattern influences the rate of crop development (phenology). Particularly important is the flowering stage which should occur after the frosts but before the onset of heat stress days.
• the relation between water supply and the evaporation affects plant growth, but it can also affect the spread of disease and insects.

The success of any crop or pasture depends on how its life cycle fits into the climatic restraints of the region.

b. Soil

This provides the base for agriculture; the important factors are depth, texture,-because it influences water holding capacity, erodibility and nutrient supply.

c. Farmer’s time.

At certain times of the year, the farmer is fully committed with his operations. Any research recommendation that requires a farmer to do something extra at these times is unlikely to be adopted. The capriciousness of the weather also affects the timing of the farmer’s day to day programme.

Farmers generally are looking for research findings that can fit easily within their environment and give economic benefits on the farm, either through higher yields or reduced costs. Too often the benefits of research are associated with the elegance of the research or the amount of money spent on it.

What is needed in the agricultural arena i~ a number of markers by which the farmers and the research officers can evaluate the benefits of research on the farm. Some of the more important markers are:-

a. Potential yield values

These are a measure of the efficient use of sunlight. A farmer’s yield can be compared with the potential yield which is calculated from a relation between production per unit of water and the average daily pan evaporation from sowing to harvest. Relations can be derived for different crops and the difference between actual and potential yield determined (Figure 1).

b. The production per unit of rainfall

This is a simpler marker in which maximum yield per unit of rainfall during the growing season (e.g. April-October) has been defined. Differences between actual and potential yield can be related to factors such as inadequate fertilizers, delayed time of sowing, lack of disease control etc.

c. Soil values

• Any loss of soil through erosion will affect future productivity.
• The available soil water content at sowing has a large effect on yield, for not often does the rainfall from sowing to harvest reach the optimum amount for crops. An available water content of 100-150 mm at sowing gives a high degree of drought insurance and tillage practice, whether long fallow or direct drilling, should aim at accumulating this amount of water.
• Upper and lower levels of soil fertility e.g. carbon and nitrogen content, need to be defined for all soils. It is not an appropriate management strategy to try and maintain a soil value at a high stable level. Soil values will fall when the exploitative factors dominate and rise again with restorative practices. It is essential to define the minimum value below which the fertility should not fall (Figure 2). Above this value, a variety of rotations can be practised successfully.

d. Yield trends

It is not unreasonable to expect the benefits of research to show up in better farm yields. Evidence from New South Wales (Bond and Ole 1980),Victoria (O’Brien 1982) and South Australia (Cornish et al 1980) however shows that the district yield of wheat has not increased over the last 20 years. It is also apparent that the variability in yield from year to year has increased, suggesting a lack of ability to control some limiting factors. Some practices, while solving one form of degradation, may generate a new degradation cycle e.g. herbicides may create chemical residue problems, and fertilizers intensify acidity problems. The variability suggests also that new technology may increase the dependence of crops and pasture on the weather in the growing season.

How then are we going to try and overcome the present limitations of translating research findings into on-farm benefits?

Guidelines for the Future

No one set of experiments solves agricultural, problems once and for all, and there will continue to be extensive research and investigation using the traditional methods of research. However an increasing problem in the future will be the need to synthesise research findings into on-farm practices and to reduce the time of adoption by farmers. Perhaps a quarter of the research effort should be allocated to this on-farm research and the selection of the priorities should be made in conjunction with representatives of farmer organisations. Some examples of the type of on-farm research are as follows:-

a. Soil Conservation

The main emphasis should be to carry out contour banking and sand stabilisation work programmes in conjunction with District Soil Conservation Boards and farmer groups. Reducing erosion and soil losses is urgent, there is no need to conduct time consuming and costly experiments to set tolerable levels of soil loss. In most cases we can’t afford any further loss of soil. Protection should be available at all times, for most of the soil loss occurs with infrequent storms. Thus at Wagga, 84% of the soil loss occurred in storms in only 5 years out of 22 (Adamson 1974). Soil erosion not only reduces the productivity base on the farm, but causes off-farm damage to roads, towns and reservoirs. These effects are costly and should be prevented. The prime focus of any erosion research should be to devise strategies that give adequate cover to the soil at all times of the year either with stubble, dried annual pasture or perennial plants.

b. Cropping

While many experiments have evaluated the effect of one or two inputs on crop yields, there is a need for field experiments which incorporate a large number of research findings into farming practice. A good example is the development of a 3 Tonne Club among farmers in Western Australia. Here, the farmers and the research officers identified 23 factors which affected yield and these were tested in combination in a cropping system on 30 hectares on each farm. Measurements were made of the soil water content at sowing, the dry matter at flowering and the number of grains per square metre, and actual yields were compared with the calculated potential yield.

Another need is for some lateral thinking into the duration of cropping. Traditionally, crops have been grown in rotation with leguminous pastures and much research has shown the essentiality of pastures to maintain the nitrogen supply. This was also the situation in England over thirty years ago, when an experiment with continuous cropping with wheat first started. Yields fell rapidly during the first 5 years but since then, as a result of a different technology, yields have climbed to nearly 6 tonnes/hectare (Bullen and Jarvis 1982). In South Australia, one farmer in low rainfall sandy soil has been on a rotation of fallow-wheat-stubble for 50 years without loss of yield or soil degradation. In a current Department of Agriculture rotation trial, best production has come from continuous wheat with nitrogen fertilizer, and the yield difference between application of 80 kg ha N and 40 kg ha N has narrowed sharply after only 5 years. Perhaps our scientific resources have spent too much time polishing the old system. Scientists do have a social responsibility to see that their work is relevant to farming and it can produce on-farm benefits.

c. Pasture

Similar types of experiments should also define ways of maintaining a continuous grazing system on annual legume pastures. Over 12 years at Minnipa (annual rainfall 325 mm), sheep numbers were maintained on medic pastures at twice the district stocking rate with the input of only 30 weeks of handfeeding. In a subterranean clover grazing trial at Mt. Bryan (425 mm rainfall), increasing the stocking rate increased wool production but reduced firstly the amount of sub clover and then the amount of barley grass; and increased the amount of cluster clover.

Within either a continuous cropping or grazing experiment, it is possible to monitor the changes in water use efficiency, the changes in soil carbon, nitrogen, phosphorus and pH and the economic benefits to the farmer.

Farmers contribute to research funds through levies on wheat and wool. Some of these funds should be offered to organisations specifically to carry over and monitor on-farm research as described above.

A further need in the decision making process is to develop a greater capacity to forecast production during the growing season. The advent of computers on farms together with climatic data sets will encourage this trend. A simple example is to upgrade the estimates of yield of crops and pastures during the growing season. Thus at the end of July , yields can be related to the probability of high rainfall (decile 9) or low rainfall (decile 1) falling in the remaining months of the growing season. A better estimate could also be obtained by adding the data to a water balance, crop development model.

Australia has about 170 000 landholders who make decisions on the use and productivity of land. These landholders should be more directly involved through their farmer groups in conservation works programmes and in research and development programmes into on-farm practices. They can help select research priorities, advise government on the need for changes in policy, including the provision of incentives to encourage adoption of new practices, and thereby contribute to maintaining a productive and economic agricultural system within the variability of soils, season and costs.

References

1. Adamson, G.M. (1974) - Intern. Assoc. Hydrological Sci. Publication 113.

2. Bond, G.E. and Ole B.T. (1980) - Proc. Aust. Agron. Conf. Queensl 1980. p 245.

3. Bullen E.R. and Jarvis R.H. (1982) - J. Roy Agric. Soc. England. 143: pp 81-88.

4. Cornish E.A., French R.J. and Hill G.W. (1980) - Agric. Record 7, no 12.

5. O’Brien,L. (1982) - J. Aust. Inst. Agric. Sci. 48:163.

Figure 1: The relation between the yield of grain per millimetre of water use and the average daily evaporation from sowing to harvest for S.A. field sites. The curved line represents the potential yield for southern Australia. Top values for other sites are also shown (T Tamworth, C Coolamon, L = Leeton, Wg = Wagga, R = Rutherglen).

Figure 2: A concept illustrating how soil fertility can be monitored between upper and lower levels. The effect of different rotations causes fluctuations between the levels. Restoration practices are needed when the soil values approach the lower level.