The type of measurements that you’ll make depends very largely on the type of trial that you’re running. You’ll find specific advice on these in the “On-Farm Trial Guide – Type of Trial” Booklets.

Monitoring your trial regularly will enable you to catch and record both kinds of crop events, which will help to interpret the final trial results. It’s a good idea to have a look at your trial at least once a month. If your visual observations alert you to interesting or useful treatment differences, it may well be worth making objective measurements of them.

MORE

Observation of pre-planting treatment effects is likely to be most important in situations where different starting conditions form part of the treatments.

Details of the types of pre-planting observations that might be made in a fertiliser trial are given in the fertiliser rate and fertiliser method trial templates. Additional or different factors may be important to you. If that is the case, devise measurements that enable you to quantify how those factors differ and what impact that might have on crop performance.

If you are devising pre-planting measurements that aren’t covered in a trial template these general guidelines might help…

Look at how your two treatments differ.

• Which differences are likely to have an impact on crop establishment or performance?
• Can you measure those differences directly by counting or weighing?
• If not, can you measure them indirectly using a scoring system or proxy variable?

Proxy variables can be very useful if they integrate a range of properties. The number of days of surface ponding after heavy rain (for instance) can be a good proxy for soil structure. It incorporates elements of aggregate stability, infiltration and permeability that, on their own, aren’t the best indicators of soil structure.

While it is important to know whether the treatments were applied as planned, it is even more important to know whether the treatments eventuated as planned.

Because lots of things can happen to a crop between sowing and harvest it’s important that you check that you’ve actually achieved the treatment differences that you set out to achieve. Wherever possible, make direct measurements of fundamental treatment differences.

This is a simple task where you can readily measure the treatments themselves. Unfortunately, it is not always possible to directly check that the treatment has eventuated as planned. It isn’t possible, for instance, to measure the amount of fertiliser that has been added to a paddock once it’s been applied! For this (very common) type of treatment, it’s important that close observation of other factors is made during the season.

The key to really successful trial monitoring is to keep your treatments in mind while you’re looking, and seek things that may or should differ between them.

If your visual observations alert you to interesting or useful treatment differences it may well be worth making objective measurements of them. If you think there have been growth differences early in the season, why not harvest some plants and weigh them to see how different they really are? This will enable you to confirm or deny your suspicions and will provide ‘hard’ numbers upon which sound practical and financial decisions can be based.

It’s also important to anticipate and look for both direct and indirect effects of the treatments applied. Direct effects are those that are a straight-line consequence of the treatment that you imposed (eg. different plant populations in a plant population trial). Indirect effects are those that are a secondary response to the treatments (eg. increased lodging due to high plant population).

Examples of important indirect treatment effects include things like…

• The simple act of cultivating soil differently can have significant effects on fertiliser requirements, crop emergence, crop development and pest and disease incidence, quite apart from its effect on soil tilth.
• Broadcasting fertiliser can increase weed vigour compared with banding, by enhancing weed access to nutrients.
• Side-dressing fertiliser can reduce plant populations (‘Sheffield canker’) and growth (root pruning) if placement is too close to the row.

It’s worth devoting some time to anticipating indirect treatment effects because, even though they aren’t always obvious, they can often swing the balance for or against a given treatment.

For example, a trial examining the effect of fertiliser placement on yield may have shown that broadcast fertiliser increased yield more than fertiliser applied ‘down the spout’. That might be the real result, or it could be an artefact of the trial method. It’s possible, for instance, that ‘down the spout’ application had a lower yield because it had a lower population. Without your knowing it, the high concentration of banded fertiliser generated ammonia gas and, because it was close to the seed, it killed quite a few seedlings. Only by measuring emergence and final plant stand would you have found out that this had occurred.

End of section critical decision point

Have you been monitoring your trial regularly? If not, will you have learned anything about why your final results turned out as they did? Have you made measurements that show that your treatments actually turned out the way that you planned them? If not, can you be sure that your trial is comparing what you think it is comparing? Have you made measurements to verify that the application of treatments hasn’t had unintentional or indirect effects on crop performance? If not, can you be certain that you’ll draw the correct conclusions from the trial?

If you can’t give positive answers to the above questions, you may want to treat your trial results with some caution – you just can’t be certain that they are reliable. If, on the other hand, it all looks OK, then move on to the next section – “Finishing the trial – getting the harvest done”.