THE CLIMATE KELPIE BLOG: Farming with more than intuition

Posted by BCG on 29th March 2022

South Australian Port Germein farmer Barry Mudge remembers the moment he decided he wanted to farm with more than just intuition. “My dad Laurie made a lot of money on the farm in ’78- we had the best yields we had ever achieved. But he’d only put in 370 acres of wheat. My question was why hadn’t we planted more? This propelled me to start searching for a way to ‘pick the year’.”

Barry Mudge

In 2006 Barry, who is also a consultant farm economist, was awarded the South Australian Farm Innovation Award for his Mudge Index, essentially a tool which appeared to in fact, pick the season.

“It had picked six out of eight years using forecasts but in the end I think it was a fluke so my search continued,” Barry said.

Rapid Climate Decision Analysis (RCDA)

Barry, whose farm lies outside of Goyder’s line (highly variable rainfall and soil type), started working with SARDI Climate Applications Principal Scientist Dr Peter Hayman to develop the tool Rapid Climate Decision Analysis (RCDA) to, as Bill Malcolm, Professor, The University of Melbourne says “Imagine the future with rigor”.

“As farmers, we are always interested in the forecast but I wanted to try to turn that interest into greater value. And forecasts are only ever a small part of the picture—a lot of what we do on the farm we are going to do anyway as it’s best practice.”

For example, the extreme weather risk Barry has on farm is heat during spring. Therefore he and his son Jonno will always sow early and maximise plant available water to try and avoid yield penalties from this heat. This decision will not change based on the forecast: “The last few wet springs predicted which have eventuated have missed us anyway so we don’t base our sowing time on the forecasts,” Barry said.

“It’s determining what decision are we going to actually change based on a forecast we have some faith in [e.g. spreading urea] that the value in forecast lie, that helps us realise opportunities/pick the year.

“The other factors in imagining the future with rigor are how reliable are the forecasts and what is the risk profile for each decision.”

FWFA quintile forecasts

“I like the output of the new FWFA quintile forecasts because they are more specific,” Barry said. Barry uses this new FWFA tool for one-to-three-month outlooks and the BoM Access charts and ECMWF for shorter term weather forecasts.

“The FWFA quintile forecast’s skill level improves as we move through the season, so while the accuracy is low at sowing time and we are less reliant on it in making a decisions at this time, when we reach July, particularly if some of the stronger climate drivers are in play, the forecasts hold more weight and we are more confident in using them to sway decisions such as urea and fungicide application come spring,” Barry said.

Barry also values the way the FWFA quintile forecasts and the FWFA suite of products have been developed: with farmers for farmers. “I was involved in the testing of the FWFA quintile forecast. There were a lot of great features in the experimental stage that I used every day. It is a positive step forward that farmers are involved in the testing of these products in effort to ensure they have value for farmers, in a readily accessible and easy to use platform.”

Figure 1: Decile bars. Rainfall forecasts for Wagga Wagga for upcoming months and seasons (generated on 30 August 2021). The forecasts show the probabilities across five different decile ranges. The long-term average probability (“usual chance”) for each category is 20% and the forecasts show the shift in the odds compared to usual. For example, for October, the odds are

How Barry uses RCDA and FWFA quintile forecasts

“When I have a decision that warrants it e.g. applying urea, buying sheep etc this is when I use Rapid Climate Decision Analysis. Simply speaking the tool is really just imagining the future broken down into five possible outcome and developing a partial budget, for example using grain prices, fertiliser prices, feed availability on each.”

The five scenarios Barry imagines are:

  • Really poor
  • Below average
  • Average
  • Above average
  • Really good

At this point in time each outcome is evenly weighted at 20 percent. Once Barry has analysed the situation he then use the FWFA quintile forecast which then skews the probabilities one way or the other.

“For example if I was going to buy sheep but the timing of the purchase depended on the forecast I would use RCDA for both an early purchase and a later purchase. While the early purchase may mean more wool I’d have to feed them if it were dry. If it were dry, it might mean that sheep would be cheaper later. Going through the simple RCDA process gives you a clearer picture of all outcomes. I can then use the FWFA Quintile forecast to lean one way or the other, being able to make a more informed decision than if I were just using the forecast and intuition.

Figure 2: An example of the graphs Barry generates through using the RAPID tool. These ones are comparing the effect of changes in seasonal outlooks on the range of outcomes from choosing to apply urea or not.

“It’s really a way of taking on risk, measuring and confronting risk rather than being afraid of it, before you actually take it. Doing this we can start to realise opportunities and load the odds in our favour before we spend a cent.

“This can be done in Excel or just a bit of paper and only takes around 20 to 30 minutes. The important thing is that you do it. Slow down your thoughts and analyse the situation. So often the numbers come out significantly different to what my intuition was telling me.”

The tool also helps to iron out bias. “We have a natural bias to try to avoid loss. Losses tend to hurt more than the benefit of an equal gain. We don’t want to lose but with this simple tool we can actually see both sides and therefore see more equally what we have to gain (and lose).”

Wider industry use of FWFA quintile forecasts and RCDA

The FWFA quintile forecasts Barry believes, could be used by anyone who is actually going to change their decision based on, at least in part, the forecast: “The FWFA Quintile forecasts also show minimum and maximum temperatures. They are not just for rainfall. This information could be used by a farmer in any industry that is in a frost prone region producing a frost prone commodity, so the grains industry but also livestock, viticulture and horticulture. We predominately use it for maximum temperature as we are more affected by extreme heat rather than frost during spring.”

Barry believes RCDA could also be adopted by any of the agriculture industries that need to make climate sensitive decisions. “The most important aspect of the tool is that it provides a framework to understand the risk profile of the decision in the absence of any forecast – then it is a relatively easy extension to apply some forecasts if you have faith in them. The output from the FWFA quintile forecasts is excellent in that it shows changes in the probabilities compared with historic data, but there are still no guarantees.”

While Barry has found the RCDA approach useful, he says it clearly cannot be used for every decision: “It will only be a few occasions and with decisions involving relatively high stakes, that it might be appropriate to slow down and invest some time in the process.”

More Information:

Barry Mudge 0417 826 790

For more information on the Rapid Climate Decision Analysis watch:

Leave a Reply