Very interesting question . You develop critical level of soil phosphorous by any method, either by creating the gradient or under typical farm conditions , it matters very little, as long as such critical limits are cross validated in a variety of soil conditions and crops for which these limits are intended. Reproducibility of such soil test -crop response results is of huge importance , needless to say , a soil critical limit can never be valid under different soils of different mineralogy or land uses. This issue needs to be addressed through rigorous modification in our test results accruing through different filed experiments with variety of land uses ....Ironically , this is where , we are lacking the most , even now...

I think on farm is the better approach as it is very difficult to maintain fixed plot gradient.

The selection of the most appropriate soil-testing procedure for a particular situation should be based on a available information relating to soil chemistry and mineralogy, and on the form of P fertilizer to be used. Conducting on-farm calibration research with different crops across several site-years can provide the basis for the interpretations for a particular soil P test. So, conducting such studies enables researchers to find the fields with different P levels. Overall, creating the gradient through depletion of P is very difficult in many ways.

Greetings Getachew,

That is an interesting question and I feel sure you will receive many very different suggestions. Mine is probably one such.

I was involved in soil test calibration studies in KwaZulu-Natal for over 30 years and, of course, faced the same problem as you do.The ideal is long-term experiments on research stations. However, this inevitably means that you don't cover the full array of important soils. We eventually decided to make extensive use of farmer-field studies, but made sure at the outset that we could guarantee a high level of co-operation. Essentially, we hired the land involved, but luckily were never required to pay any money, as progressive farmers realized that they would benefit from the results. The areas, which varied between a ha or so and 12 ha, where several different experiments were conducted, were marked out with permanent concrete corner markers. 30-40 cm of flexible steel cable was precisely embedded in the concrete so as not to disturb tillage operations and we were able to accurately run trials for up to 25 years. Initially, farmers did the tillage for us, but in latter years we transported a tractor, plough and planter to each site. The farmers did not have to concern themselves with the running of the experiments, but provided the land on a semi-permanent basis. As you are aware, long-term experiments provide extremely valuable research vehicles.

I trust that this approach is of some use to you.

Yours sincerely,

Mart Farina.

Scientifically fixed plot by creating a gradient is the best approach. Though it is time taking process, but the final results which we are getting in this approach is more reliable than on-farm approach. So in my opinion fixed plot is the best approach.

Thank you so much Dr. Farina for sharing us your accumulated experience in the field. As you know the first set of soil-test P crop response trials were conducted under on-farm condition on selected soil types using major crops in the country. Some interesting results were produced, published and validated on farmers fields. Since the last four years the approach has been changed to permanent plots and trials have been carried out experimental stations. But, the problem here is that, as mentioned above, creating the gradient to apply different levels of P has been difficult because of high native P level in some soils.

I hope your contribution in this regard will continue.

Dear Getachew,

Thank you so much for your response. As you may know, my friendship with Girma Gebre Kiden, Prof. Takaligne Mamo, Dr. Bataino and others has been a privilege for me. So, too, your lovely country. My email is [email protected] and I am open to any comments or questions you may have. Keep in mind, though, that I have been out of active research for some 20 years. Many things have probably changed in that time?

Keep well and kind regards,

Mart.

Very interesting question . You develop critical level of soil phosphorous by any method, either by creating the gradient or under typical farm conditions , it matters very little, as long as such critical limits are cross validated in a variety of soil conditions and crops for which these limits are intended. Reproducibility of such soil test -crop response results is of huge importance , needless to say , a soil critical limit can never be valid under different soils of different mineralogy or land uses. This issue needs to be addressed through rigorous modification in our test results accruing through different filed experiments with variety of land uses ....Ironically , this is where , we are lacking the most , even now...

I fully agree with the views of Dr. Srivastava, we need large number of field experiments under variable soils and crop conditions to produce high quality of soil test crop response data which is still lacking at present.

I agree with Dr. Anoop, and you raised a very important point " Reproducibility of such soil test -crop response results is of huge importance , needless to say , a soil critical limit can never be valid under different soils of different mineralogy or land uses. "

My follow-up question is that what should be the P levels of fields when they are selected for soil P test studies to determine P critical concentrations? In fact, many of the fields selected for such studies need to have low and medium P levels, with some high P levels.

Dr Getachew, the critical soil test level for P will largely depend upon method of extraction for P , soil pH, calcreousness , mineralogyetc etc , besides type of test crop to envisaged and the level of crop yield intended to raise from the current level to anticipated yield level...A soil test value optimum for one yield level will become sub-optimum for higher yield level. Such crop-based soil test interpretation , we need to develop , then only such relationship will be somewhere near to original field conditions....