THE natural instinct of farmers is more valuable than climate prediction tools when it comes to yield forecasting and managing seasonal risk, according to Mingenew-Irwin group manager Cameron Weeks.
³The large variation in seasonal rainfall means that farmers need to be able to manage these fluctuations so that they can capitalise on the good seasons and minimise the damage in the poor seasons,² he said.
Mr Weeks is coordinating the Horses for Courses project, investigating a range of decision-making support tools available to farmers for in-season yield forecasting.
The project looks at the value of using climate prediction tools such as APSIM, SYN and PYCAL, when compared to using historical data as a trend predictor.
³With regard to utilising the information from yield forecasting, nitrogen is the major crop input that can be varied once the crop has been sown, with the required nitrogen rate largely determined by the crop yield potential,² Mr Weeks said.
SYN and NULogic overestimated nitrogen requirements in 2004 on the test farms.
But Mr Weeks said leading agronomists in the region also overestimated nitrogen requirements in 2004.
³Really this means that none of the yield forecasting tools available, nor the experience of agronomists could predict the season in 2004,² he said.
³From this trial of SYN, PYCAL and APSIM, I can not see why a farmer would pay money for a yield forecast they could figure it out themselves using historical rainfall data.²
Mr Weeks believed playing the season was a key to better managing seasonal risk.
He said to play the season, crop yield needed to be forecast at numerous times throughout the season.
The basic principles of yield forecasting were, understanding stored plant available water at a given point in time, estimating likely rain to come and reading the crop.
With a good understanding of these three factors it was possible for a sound estimate of likely yield to be made without the aid of a decision support tool.
Following on from the excellent season in 2003, 2004 was a mixed bag for farmers in the Mingenew-Irwin area.
In terms of rainfall 2004 was poor.
Totals for the year across the district were well below the long term average, with all sites in the Horses for Courses trial listed as experiencing decile 1 to decile 3 years, that is, when compared with all seasons on record they were in the worst 10-30pc ever experienced.
Luckily the area received an excellent September rainfall which was in the best 10-30pc experienced, making September a decile seven to nine month.
In terms of yield, this generally meant below average returns, except for much of the sandplain and coastal areas, where totals received were almost ideal.
The positive from this lower rainfall season comes from looking closely at the rain received relative to the yields achieved.
Farmers in the Horses for Courses trial achieved water used efficiencies of 80-100pc, meaning that farmers achieved close to the maximum yield possible with the rainfall received.
This is a credit to their farming skills and their ability to play the season.
³I believe that decile information is the most powerful tool available to farmers at this point in time,² Mr Weeks said.
³There is a strong correlation between rain at the end of one month and rainfall in the subsequent month.
³For example, in most years, farmers in the Mingenew-Irwin area receive two thirds of their growing season rainfall between April and July and the remaining third from August to October.
³This historical trend gives farmers an indication that if the season has been poor between April and July, then it is most likely that rainfall will continue to be low from August to October.
³The best that a farmer can do is play the season.
³For example if there is good rain at the break of season, apply fertiliser accordingly, then reassess two weeks and four weeks after sowing.
³If the trend changes, then change yield forecasts and subsequent fertiliser decisions accordingly.²