How risky is your rotation?
Author: Roger Lawes, CSIRO | Date: 24 Feb 2015
Key messages
- Agriculture is risky, but appropriate crop rotation can stack the odds in your favour.
- An economic simulation for a sand over gravel at Kojonup in Western Australia demonstrates how a loss making pasture can generate future profits through higher yields in subsequent wheat crops.
- The higher wheat yields occur because soil borne diseases and weeds are controlled during the pasture phase.
- The analysis demonstrates that the agronomic package chosen in one year influences future costs, returns and risks for future crops.
- Break crops help maximise profits in better years and lower overall risk, if managed well.
Aims
In a business, risk is discussed with reference to the potential reward, where more risky ventures offer larger returns than more conservative approaches. Agriculture is a risky enterprise and the risk is driven by climate and commodity price, where climate influences crop yield.
The effect of climate on yield potential varies with crop type, as cereals are more resilient to drought and water logging than pulses or oilseeds. Climate also influences the expression of problems like soil borne disease. Climate will also influence the seed set by weeds in much the same way that climate influences grain number and grain size in cereal crops.
Therefore the interplay between climate, crop choice and extent of biotic stress complicate any assessment of risk in agriculture. The complication arises because management actions in one year contribute to the risks in the following year. For example, a high yielding season for a cereal crop will deplete nutrient reserves and these nutrient reserves will need to be replenished in the subsequent crop. Similarly, wet winters create an environment that helps disease inoculum build up, and if there is a dry summer, diseases such as Crown Rot and Rhizoctonia are more likely to carry over into the following year. If any weeds escaped during a wet season, they will produce vast numbers of seeds and create a large residual seed bank that will need to be managed in the following crop. If these conditions are not managed, the following crop will be established into a field with poor nutrient reserves, a high likelihood of disease and a high weed seedbank.
In effect the agronomic practices used in one year will influence the biotic stresses and nutrient requirements of the crop for the following year. These could influence the yield of the following crop and alter the amount of inputs required to grow the crop. Consequently, even if it is a good season in the following year, the crop may not reach yield potential because the paddock was not ‘set up’ to yield well.
Risk can be explored over one year, for the single enterprise, or over many years, where the vagaries of season and the onset of biotic stress can alter crop yields and change the perception of what risk entails. It may be thought about in the context of a single enterprise in a single year, or thought about at a paddock or farm level over an extended period of time.
Over time, a crop sequence may be viewed as an intensively cropped monoculture (continuous wheat), intensively cropped with one a break crop (like canola, wheat, wheat) or farmed with less intensity with a managed pasture, followed by two crops (like pasture, wheat, wheat).
Therefore our aims are to use the APSIM crop model and the LUSO economic model to explore how the economic risk of growing a wheat crop changes between an intensive continuous wheat rotation, and intensive canola- wheat-wheat rotation and a less intensive Pasture-Wheat-Wheat (PWW) rotation for Kojonup on a sand over gravel soil.
Method
Here we use the APSIM crop model and the LUSO economic model for crop rotation to explore how the economic risk of growing a wheat crop changes between an intensive continuous wheat rotation, and intensive canola- wheat-wheat rotation and a less intensive Pasture-Wheat-Wheat (PWW) rotation for Kojonup on a sand over gravel soil.
APSIM was used to generate 100 years of wheat, canola and pasture yields for Kojonup in Western Australia. The pasture was assumed to be a legume (sub clover) pasture, with a yield distribution similar to wheat, where high yielding wheat seasons generated a high yielding pasture, in relative terms.
The LUSO model predicts the impact each crop has on the weed and disease populations that may then affect the yield of the following crop. This means that with some crop rotation choices, yields of future crops could be reduced. We compare the risk by exploring the number of seasons that exceed a profit of $0/ha for an intensive Wheat-Wheat-Wheat (WWW) rotation, an intensive Canola-Wheat-Wheat (CWW) rotation and a less intensive Pasture-Wheat-Wheat rotation. The objective here is to focus on the probability of exceeding $0/ha, and to explore whether less intensive rotations, with fewer crops and a low profit pasture can generate profits equivalent to a intensively cropped rotations with the same or similar level of risk over a 3 year period.
Each simulation was repeated 5000 times using different combination of seasons to explore the average effect crop rotation has on profit, and the effect crop rotation has on profit in different seasons. Therefore, one scenario may choose the years 1960, 1961 and 1954. The next simulation may choose the years 1957, 1981 and 1998. Each combination will produce different potential wheat yields, with different nitrogen requirements. If disease is present, the impact of that disease will also vary with season.
Results
The chance of generating a profit in the first year of the sequence was 83% for the (CWW), 55% for the (PWW) and 73% for (WWW).
In the second year wheat was grown in every crop sequence. However, the chance of generating a profit changed to 76%, 82% and 68% for the CWW, PWW and WWW crop rotations respectively. The biotic stresses and amount of nitrogen required to grow the wheat crop were influenced by the previous crop, and these stresses, if present, reduced yields and the increased nitrogen requirement of the wheat crop. Therefore the first crop could increase costs of the second wheat crop and depress yields. Collectively these two factors reduced the probability that the wheat crop would generate a profit, and this explains why the wheat crop following a managed pasture, which fixed nitrogen, controlled weeds and managed disease, was most likely to generate a profit.
In the third year, where wheat crops were again grown across all crop sequences, the chance of generating a profit declined. For CWW and PWW the probability was 69 and 70% respectively. For WWW it was just 61%, which is quite low, as 4 years out of 10, this system would incur a loss in the third year.
The changes in probability of a wheat crop generating a profit demonstrate an important fact about probability in cropping. The probabilities are different to tossing a coin three times because the previous crop does influence the outcome.
Furthermore, crop rotations run over many years. By combining the probabilities, it is possible to explore whether the farming system as a whole breaks even or makes a profit. After 3 years the probability that CWW generated a profit was 86%. For PWW the probability was 84%. For WWW the probability declined to just 74%. These differences continued when the probability of making more than $500/ha after three years was considered. For CWW and PWW this was 55% and 50% respectively. For WWW this was just 24%. Therefore crop rotation with break crops was more likely to break even after 3 years and more likely to be extremely profitable (Figure 1).
Conclusion
This theoretical analysis of economic risk for a farm in Kojonup illustrates that the type of crop rotation can influence the risk profile of a paddock. However the outcome of the theoretical model needs to be contextualised with regard to management. That is, how do you apply this analysis in a management sense? It is one thing to calculate the odds with a computer, but it is another to explore the concept in reality. The manager must ask themselves what the odds are that their system generates a profit, and what are the odds that each enterprise in your system individually generates a profit? This varies from farmer to farmer, and is an expression of agronomic skill and environment for each enterprise. Some farmers may be able to grow productive and profitable pastures. Others may be well versed in growing weed free and high yielding Canola crops. The odds must be explored in terms of the vagaries of season, the commodity price and the manager’s ability to realise the potential yield of each enterprise. These issues must be reconciled against other components of the farm business, such as the size of the businesses fixed and variable costs.
Despite this complexity, it is vitally important to load the dice in your favour, and push those probabilities of making a profit as high as possible. The probabilities are enhanced by understanding how the agronomic practices in one season will influence the agronomic and economic outcome of the following crop.
Figure 1. Histograms of cumulative profit over 3 years from a Canola Wheat Wheat (CWW), Pasture Wheat Wheat (PWW) and Wheat Wheat Wheat (WWW), rotation, as calculated from the LUSO model for Kojonup in WA. The red line indicates a profit of $500/ha.
Crop sequences take on many forms and often increase profits in future wheat crops. Importantly they appear to be most beneficial in good years, where problems like disease create large yield differences between crops following a break and crops following a cereal. Thus break crops can help maximise the upside in favourable seasons. However, it is vital that pasture breaks are well managed to ensure you generate a nitrogen benefit and reduce the risk of diseases and weeds in subsequent cereal crops. Finally, the economics of growing a break crop need to be evaluated across a number of years. This may demonstrate that a loss in one season can generate substantial future profits.
GRDC Project Code: CSA00029,
Was this page helpful?
YOUR FEEDBACK