Friday, March 29, 2024

Soil tests – there’s no one-size-fits-all

Avatar photo
Last month, in discussing reducing fertiliser applications, I cautioned against taking recommendations from anyone who can’t lift their head above the computer screen. I have been challenged on this, with the sentiment that this could be construed as a denigration of the efficacy of the whole soil-testing regime. For anyone interpreting my comment in this way I can only say, “Good on you – you’ve got it right.”
Reading Time: 5 minutes

Soil testing is far from an exact science. It is important to realise that the aim of a soil test is not to measure the total amount of a nutrient element in the soil. If this was the case soil testing would be a fairly simple and accurate procedure. But it’s not.
The aim of soil testing is to give a measure of the soil’s ability to provide essential nutrients to the plant. The problem is major nutrients are locked up in the soil in forms that are totally unavailable to the plant – so getting a measure of the total nutrient present would be a complete waste of time.
To get a measure of the nutrients available to the plant a test is required that will, in effect, mimic the plant’s action and ability to extract the nutrients from the soil. The accuracy of a test to achieve this is obviously going to be highly dependent on the chemical nature of the soil, so a test that is most suitable and appropriate for a particular soil may be much less so for another – there is no one-size-fits-all test. Extensive research and testing in the 1960s and 70s established that, across the board, the Olsen P test was the one best-suited to New Zealand soils and farming practices – but it is still not ideal or perfect in all circumstances.
In addition to variations because of soil type, there are many other factors influencing soil-test results, such as time of year, previous topdressing history, climatic factors, soil organic matter and moisture status. Then we have laboratory errors and inaccuracies, and most significantly, sampling errors. This latter is the one most likely to give big variations in soil-test levels.
Strict protocols are in place when soil sampling – all aimed at minimising sampling error. If these are not fully observed the results could well be useless at best and dangerous at worst. The best that could ever be hoped for would be a confidence level of +/- 10%, and +/- 40% is quite possible.
At the Demonstration Farm we have a very intensive soil-testing regime with paddocks being tested annually. Despite the sampling protocols being strictly observed we can easily have a five-point yearly difference in Olsen P on the same paddock.
Now, to lighten things up a bit and as a departure from the normal routine, I will relate a few anecdotes from my experiences over the year, to help illustrate and reinforce the points made.
First, I go back to the 1960s when I began working with the then Department of Agriculture as a farm advisory officer. I was very fortunate to be given a training and controlling officer by the name of Jim Graham – a man with widespread knowledge and experience and generally recognised as one of the best trainers in the department.
Jim was a fairly gruff and acerbic character, who generally didn’t suffer fools gladly. As the first step in any training programme, Jim saw it as his duty to knock any semblance of arrogance or pretensions to omnipotence and knowing-it-all from any freshly minted graduate entering the service – at which, I might add, he was remarkably successful.
One of his strictures was nobody should give an interpretation and recommendation from a soil-test result unless they had done the soil-sampling themselves. This was somewhat against the trend in thinking at the time with regard to efficient use of staff – that highly qualified, highly paid, professional people should not be expected to do work that could be carried out by a lower-paid technician.
Jim’s point was that a soil test was only one tool in the kitbag of providing fertiliser recommendations and any good adviser should be adept at identifying the fertility status of any pasture by visual inspection – and paying due regard to this when formulating an appropriate fertiliser policy.
Jim was fairly cynical about the reliability of soil-test results, in particular the possible influence of human error. Hence his insistence that all recommended sampling protocols and rules be strictly observed. He was also adept at the ability to deliver the odd pithy little aphorism appropriate to the occasion – in the manner of Steve Hansen.
The one I remember vividly was the rather slanderous statement that “those girls in the lab supposedly analysing soil samples spend most of their time talking or dreaming about what they did last Saturday night and who with, rather than concentrating on the job in hand”. He was also the author of the comment I reported last month that using a certain liquid fertiliser was an expensive way of buying 44-gallon drums.
Another of his one-liners I recall was, when presented with a soil-test result from an extremely dubious source and the triumphant exclamation of “look I’ve finally found out what the problem with my pastures is – it’s an iron deficiency” Jim’s response was “Yeah – in the form of a plough”.
Largely under Jim’s wise tutelage, I became reasonably adept at assessing the state of health of pastures and diagnosing problems purely by visual inspection.
Using this skill in my early days, in the hill-country area I was working in at the time, I became convinced that a potash deficiency was developing in the area. Thinking it through, this was quite a reasonable assumption. In the area, the introduction and widespread use of aerial topdressing of super phosphate had given a marked improvement in pasture growth and quality over the previous 20-odd years.
The accompanying intensification of farming had consequently further depleted the already meagre soil potash levels to the extent that potash was, in many cases, now becoming the limiting factor to plant growth. However, being a fresh-faced academic I found it impossible to convince the local farmers of this. As far as they were concerned superphosphate had done a great job and would continue to do so – their observation and conclusion was that super was the only fertiliser required on their soil type.
So here I am one day soil-sampling on the property of what I considered to be one of the more progressive and open-minded farmers in the district. Walking around the farm, taking samples and strictly observing all the necessary protocols, I spent the time explaining why I thought potash would be very beneficial – but I could see I wasn’t making much progress.
Completing the circuit of the farm we came to the woolshed and environs, and there was a small holding paddock with lush deep-green ryegrass-white clover contrasting vividly to the browntop-predominant anaemic-looking pasture typical of the rest of the farm.
This was my cue. “Look, we all know animal excreta is rich in potash. Tell you what, I’ll take an extra sample at no charge and we will see what the difference in potash levels are.”
The farmer readily agreed, so off we went and, unknowing to him, I tossed all the sampling protocols out the window and plonked the auger down into every urine spot I could find. I then sent the samples off to the laboratory and sat back to confidently await the results.
Ever had egg on your face?
The results came back practically identical to those in all the other paddocks tested.
I have two possible explanations.
One was that the lab girl had been following Jim’s dictum on her Saturday night activities and suddenly woke up to the fact that one result was completely different to all the others. Convincing herself that she had made a major slip-up in the analysis procedure she then corrected the supposed error by reporting figures similar to all the others.
The other possibility was that the whole soil-testing procedure was so badly flawed as to be totally unreliable.
I will leave it to the reader to decide which is the most likely explanation. Whichever is chosen, the conclusion must still be that soil-test results are not set in concrete. So that’s one anecdote – more next month.

Total
0
Shares
People are also reading