Department of Primary Industries and Regional Development
Testing of pasture soils from 2009 to 2018, as part of DPIRD’s Whole Farm Nutrient Mapping (WFNM) project in the coastal catchments of south-west Western Australia (WA), indicated that soil pH was so low it could be limiting plant access to nutrients. Observations by some farmers who had been involved in the soil testing were that lime application had not increased pasture production, even when pHCa (pH measured in calcium chloride) tests indicated that phosphorus (P) should become more available by increasing soil pH. Farmers also wanted to know if they needed to apply more P than soil testing suggested to overcome effects that soil acidity might have on reducing P uptake by pasture plants.
We hypothesised that the acid soils were more than adequately fertile in P, despite their low pH, and subterranean clover (Trifolium subterraneum) would still have access to enough P. If pastures can take up sufficient P, farmers might not need to apply lime to such soils to ensure P availability to pastures. Nor would they need to apply more P to overcome the effect of pH on P uptake. Poorly utilised P in the coastal catchments is transported in run-off and contributes to algal blooms in watercourses and estuaries, and unnecessary P application could increase input costs and the amount of P in run-off.
About 80% of tested paddocks had a pHCa less than 5.5. As pHCa falls below 5.5, the ability of pasture plants to take up P is likely to decrease because of a potential reduction in availability for uptake by plants. The soil testing also showed that about 80% of paddocks contained ample soil P-fertility, and about 60% contained ample soil P-fertility as well as being acidic (pHCa <5.5).
We used the results of the WFNM project to identify sites which had a range of soil P-fertility and pHCa levels, and we undertook tissue testing of clover plants to estimate the soil P-fertility required to avoid lime application.
For soils of similar P-fertility, increasing pH was associated with higher P uptake by clover, but these were not always significant (P ≤ 0.05). In soils with excessively high P-fertility, clover plants took up more P than needed for optimum production, regardless of soil pHCa. However, while most of the soils tested had higher P-fertility than necessary for clover production, there may be other constraints that are caused by soil acidity or other conditions which can limit plant yield, such as high subsoil aluminium, poor clover nodulation, waterlogging, compaction or micronutrient deficiencies.
When there was 90% more P in the soil than was required (that is, 1.9 times the required soil P-fertility), soil pHCa had little effect on the amount of P taken up by clover. Forty per cent of soil samples taken in the WFNM project had at least 90% more soil P-fertility than required, which means soil acidity is not affecting P uptake. Adding lime in these situations would only lift production if constraints other than P-fertility exist.
The results suggest there is unlikely to be any significant benefit on clover P concentration by liming soils with a pHCa of 5.0 or above and with at least 20% more soil P-fertility than required for optimum production. This does not mean that liming is not beneficial because nutrients other than P can benefit from improved soil pH. It does, however, help to explain why some farmers may not have achieved an increase in clover production by applying lime.
This report also provides a table of adjusted critical Colwell P values for soils with pHCa less than 5.0 to enable recommendations for P applications to be adjusted for pH.
Number of Pages
phosphorus, soil testing, soil pH, acid soils
Weaver, D, Summers, R, Rogers, D & Richards, P 2020, ‘The effect of soil pH on phosphorus content of clover pasture’, Resource management technical report 417, Department of Primary Industries and Regional Development, Perth.
Maps are not included as part of the complete document download. If this report contains a map, it will be available in the individual parts list below.
This file is 3.2 MB. Files over 3MB may be slow to open. For best results, right-click and select "save as..."