Crop Nutrition

Generic selectors
Exact matches only
Search in title
Search in content
Search in posts
Search in pages


How to assess new crop nutrition products

Every year, new fertilizers and other crop nutrition products come onto the market. Crop response to essential plant nutrients is well understood. What is less clear is the efficacy of other crop nutrition products. Do microbial amendments, acids, plant, and animal extracts, and other compounds also enhance nutrient use? Legislation on product efficacy is limited.

Some products like legume inoculants have undergone the test of time. But without long-term use, how do you know the product works?

Ask for evidence

Always ask for evidence to back up claims. Endorsements, testimonials, and self-promotion are not evidence. Ask for data from recent field trials on crops and soils relevant to your farming system. This data should include soil tests to make sure soil fertility is not interfering with the results.

When looking at trial data, check for:

  • Controls. Every fertilizer experiment should have a nil treatment (no added fertilizer). Without a control, there is no way to know if the new product actually did anything. Also check that the product was applied the same way every time.
  • Replication. Were the treatments repeated across the field? Replication helps account for natural variation in the field. Without replication, you can’t tell the difference between treatment effects and luck. Check there were at least four replications.
  • Randomization. The treatments should be randomized. This means they are not in the same place in each replication. For example, if all the control plots were next to each other, then all the treatment plots next to each other, the trial was not randomized.
  • Repeated. Has the trial been repeated on different soil types, in relevant regions, and on the same crop type? How many times did they repeat the experiment, and did they get the same results each time?
  • Compared statistically. A replicated trial will have an average result for each treatment and a measure of error. The error gives a range of “normal” values for that treatment. Treatments are significantly different when the error ranges don’t overlap.

Figure 2. Treatments should be randomized and replicated.

 

 

 

 

 

 

 

 

 

 

Watch this webinar with Rob Norton for more detail.

Ask yourself…

When reviewing trial data, do a logic check. Are the results clear, reasonable, and reviewed?

Results

Check:

  • if graph scales have been stretched
  • there are error bars on the graph or another indication of error
  • the least significant difference is reported.

Figure 1 Poor results presentation   

Reasonable

Do the results and interpretation seem reasonable? For example, if there was a big yield increase but very little fertilizer applied, be sceptical. The nutrients had to come from somewhere.

Reviewed

Have the results been published in a peer-reviewed journal? Higher quality journals require two to four years of trial data across multiple sites. Conference publications are not always peer-reviewed.

If you are planning on using a new product, be sceptical and ask for evidence. If something seems too good to be true, it probably is.

More

Webinar with Rob Norton on Evidence Based Agriculture

Fertilizer Australia code of practice for fertilizer description and labelling

IPNI – Evidence based agriculture

Review this article
Feedback
Share this:
Your feedback has been submitted

Leave a comment

Article Feedback