Previous PageTable Of Contents

Management responses to declining potassium fertility in Ferrosol soils

Mike Bell1, Gary Harch1, Peter Want1 and Phil Moody2

1 Plant Science/Delivery, Dept. Primary Industries and Fisheries, Kingaroy, Qld 4610. Email
Natural Resource Sciences, Dept Natural Resources & Water, Indooroopilly, Qld 4068. Email


Negative potassium (K) budgets in grain and cotton cropping systems have reduced soil K reserves and K deficiency is becoming more widespread. In addition, stratification of reserves in often dry surface soils will limit root access to K for extended periods. This paper reports results from field studies looking at crop yield-soil test K relationships for a Ferrosol, and from glasshouse studies to determine patterns of K accumulation in Ferrosols with different K status.

Crops varied markedly in the critical soil test K for grain yield, but there was no significant response to K placement in either grain or cotton seed yield. Patterns of K accumulation differed between crop species. While all crops exhibited faster relative accumulation of K than biomass, relative K accumulation rates in maize were >> than wheat or sorghum. This was related to the more indeterminate growth habit of the latter two species resulting from addition of tillers, when compared to maize. Delayed access to surface applied K until flowering in all species resulted in significant late crop K uptake but no significant biomass or grain yield responses, even under conditions of severe K deficiency.

Key Words

application method, crop uptake, grain yield, sorghum, maize, cotton


Potassium (K) is a nutrient required in large quantities by crop plants, with only nitrogen found in greater concentrations in plant biomass. However, K fertiliser inputs in Australian cropping systems have generally been low relative to other high yielding production systems, with Reuter et al. (1997) estimating that nationally Australia had a negative K balance, with that balance strongly negative in broad acre cropping regions. The negative K balances recorded for rainfed cropping systems in the Red Ferrosol (Isbell 1996) soils of the inland Burnett (Bell and Moody 2001) are consistent with the widespread evidence of declining and increasingly stratified soil K reserves, leading to K deficiency and yield losses (Bell et al. 1995).

Once K infertility is recognized as a constraint to productivity, or as soils are diagnosed as being in the later stages of declining K reserves, the key issues then become determining when to start applying soil K amendments (by the use of valid soil and plant diagnostic criteria) and how to apply those amendments most economically. In crop and pasture systems in southwest Western Australia, where sandy soils predominate and redistribution of applied K down the profile by leaching is often recorded (Edwards, 1998), surface broadcasting of K fertilisers after crop or pasture establishment is an effective way of amending K deficient soils. However, this approach is not suitable for heavier-textured soils typical of rain-fed cropping systems in Queensland and northern New South Wales. Leaching of K does not readily occur here in clay loam soils with moderate cation exchange capacities, such as the Red Ferrosols of the inland Burnett (White 2002).

Fertilisers are primarily applied by banding in these rainfed cropping systems, either prior to sowing (N), in the seeding row at planting (N, P and trace elements like Zn) or as a side dressing after crop establishment (N). However there are limited opportunities to add K fertilisers to the mix applied at planting, especially in crops grown in rows spaced 0.75m-1.5m apart, due to negative impacts on crop establishment through salt effects, so alternate application strategies are needed. White (2002) found limited responses to muriate of potash (KCl) applications when applied in deep (ca. 30cm) bands by rainfed crops on K deficient and strongly stratified Red Ferrosol soils, despite a number of crops clearly showing K deficiency symptoms, and it was hypothesised that the lack of responses were due to either (i) small soil volumes enriched by banded K applications, (ii) an inability of crops to proliferate roots in and around the KCl band or (iii) poorly synchronised K access relative to crop demands. This paper reports field and glasshouse experiments examining timing of K uptake in different crop species and the yield responses to different rates and application methods of K fertiliser on a deficient Brown Ferrosol near Kingaroy.


Glasshouse study – patterns of crop K accumulation

Pot studies were conducted in the glasshouse to determine the relative rates of accumulation of N, P, K and dry matter in sorghum and maize grown in soils with low, medium and high concentrations of exchangeable K (Kexch: 0.2, 0.35 and 0.6 cmol(+)/kg). Nutrients other than K were non limiting. Destructive samples (4 replicate pots per sample time) were collected on 5 occasions during plant growth and at maturity, with DM recorded before samples were ground and analysed for nutrient concentrations.

The experiment was repeated in a second season, with 2 variations. In the first, after the 3rd destructive sampling (taken at anthesis in all species) 4 replicate pots were top dressed with K fertiliser which was surface incorporated and washed into the top soil of the pots, and the impact of late K application on plants with differing K status was determined at maturity. The second was to separate the main stem and tillers for the sorghum and wheat plants to determine the relative accumulation of biomass and K in each.

Field study – K rate * application method

Field experiments were established on a Brown Ferrosol at Taabinga (Kexch of 0.11, 0.06, 0.04 and 0.02 cmol (+)/kg in the 10cm, 20cm, 30cm and 40cm profile layers) to investigate the interaction between K application rate and the depth and method of incorporation of K fertiliser in the soil profile on the availability of K to subsequent crops. The site was able to utilise supplementary overhead irrigation during prolonged dry periods. The trial was established as a split plot with three replicates. Main plots consisted of three depths of soil tillage/K fertiliser incorporation – (a) broadcasting onto the soil surface followed by shallow incorporation with offset discs to a depth of 10-12cm; (b) deep banding at ca. 30cm using a Yeoman ripper with tines spaced at 35cm intervals; and (c) broadcasting followed by profile inversion (using a square plough) to a depth of ca. 30cm. Subplots were K application rates (applied as KCl) equivalent to 0, 50, 100, 175, 350 and 500 kg K/ha, with residual values followed through 4 summer crop seasons. Prior to fifth (2006/07) crop season, each subplot was split to allow a supplementary K fertiliser application of 100 kg K/ha, so each plot (with contrasting residual K from the initial application) could be compared to an adjacent subplot with K fertiliser. Impacts of K fertiliser were assessed in terms of biomass accumulation and grain yield for the 2006/07 (maize and cotton) and 2007/08 (grain sorghum) summer seasons.

Results and Discussion

K rate * incorporation method

Initial significant impacts of incorporation method on distribution of applied K down the soil profile were eroded by the effect of redistribution of K to the soil surface in plant residues (Bell et al., 2008), such that the only significant differences in 2006/07 were between K rates at all depths in the soil profile (Figure 1). Exchangeable K did not differ significantly between the 2006/07 and 2007/08 seasons.

Figure 1. Effect of rate of applied K on exchangeable K in 10cm profile increments to 40cm in a Brown Ferrosol in the inland Burnett. Values are means of all incorporation methods, and vertical bars indicate lsd values (P<0.05) for each depth increment.

The clear evidence of enrichment of the deeper soil layers (especially the 30-40cm layer) at high K application rates suggests that, contrary to the findings of White (2002), there may be some leaching of applied K in these soil types. The key factor seems to be the high application rates, which may be necessary to saturate the specific K sorption sites that were speculated to occur in clay soils, or simply high enough to negate the significant redistribution of K onto the soil surface in crop biomass occurring at harvest each year.

The 2006/07 (maize and cotton) and 2007/08 (sorghum) crop seasons illustrated the quite different K requirements and critical soil exchangeable K concentrations for different crop species (Figure 2). The critical soil K needed to achieve 90% of maximum yield in the grain crops (0.19 cmol(+)/kg) was significantly lower than that recorded for cotton (0.30 cmol (+)/kg), and is consistent with the observation of K responses in cotton in an increasing number of districts (Bedrossian et al., 2000). The value for cotton may actually be even higher, as maximum yields in this trial were only 2.5 t/ha (seed + lint) due to delayed planting and use of lower yielding varieties.

Figure 2. Relative yield responses to soil exchangeable K in the top 10cm of the soil for maize, cotton and sorghum grown in the field in 2006/07 and 2007/08. Response surfaces describing the pooled response for sorghum and maize are contrasted to the response for cotton, with dashed lines indicating the exchangeable K concentration required to achieve 90% of maximum yield.

The statistical models were not improved by addition of soil K concentrations from deeper layers, either as additional factors in a statistical model or by calculating a weighted K profile for each treatment (data not shown). This may be due to the relatively strong correlation between K concentrations in different soil layers evident in these later years of the study, due to the apparent leaching evident in Fig. 1. Further studies into the relative importance of varying proportions of the soil enriched with K are planned to study this issue.

Patterns of K accumulation in pot trials

Relative rates of accumulation of both dry matter (DM) and K were similar for plants grown in soil with low, medium or high soil K concentrations, so results have been pooled for this paper. The relative rates of DM accumulation during crop growth were quite similar for sorghum and maize (Figure 3a) in the first experiment, but the patterns of K accumulation were quite different. In sorghum, K accumulation paralleled, but was slightly in advance of, DM accumulation. In contrast, K accumulation in maize occurred very rapidly, with little K uptake occurring after H3 (which was taken at anthesis in each species). This different pattern of K accumulation suggested that maize would be more susceptible to short term K deficiencies (eg. when surface soil layers were dry in a strongly stratified soil profile) than sorghum.

More detailed plant growth analysis in the second study showed a similar response in maize (data not shown) and sorghum, but when the dry matter and K accumulation by sorghum plants were separated into main stem and secondary tillers, the reason for the contrasting sorghum response became clear (Fig. 3b). In the sorghum mainstem, K accumulation occurred rapidly and well in advance of DM in the same fashion as seen in maize (Fig. 3a). However, the addition of tillers post anthesis resulted in significant proportions of both plant biomass (average of 38% of whole plant biomass at maturity) and K (average of 42% of whole plant K at maturity) accumulating in the latter part of the growing season in these parts of the plant. As the tillers were actively growing and accumulating K right through until maturity, the effect was to slow the relative K accumulation pattern on a whole plant basis.



Figure 3. Relative accumulation of dry matter and K in (a) sorghum and maize in experiment 1, and (b) in whole plants or tillers for sorghum in experiment 2. Data are the means from plants grown in all soil K levels

The indeterminate nature of plant growth in sorghum, relative to crops like maize, results in a more gradual accumulation of plant K during the growing season. However, whether this different K uptake pattern means that sorghum is less susceptible to dry top soils pre-anthesis in stratified profiles has yet to be determined. In situations where higher plant populations are sown and tillering is reduced, the differences to maize may be negligible, and further work to elucidate this is required. Both crop species were able to take up K when applied at anthesis, but uptake was only significant in low K soils where late-applied K raised whole plant K uptake by 50-80% in both species (data not shown). While plants were able to accumulate K late in the season in conditions of inadequate K status, there was no response in biomass production or grain yield.


The studies reported here provide evidence of contrasting responses to increasing soil K status between crop species, although as yet there is no evidence to suggest a more thorough soil mixing than that provided by offset discs is necessary. There is strong evidence of critical uptake periods where K uptake occurs well in advance of biomass accumulation, especially in strongly determinate species like maize, and K available late in the growing season cannot overcome yield reductions from earlier deficits – at least in grain crops like sorghum and maize. Further work is needed on other soil types (especially Vertosols) in the northern grains and cotton growing regions.


Bedrossian S, Singh B, Wright P (2000). Premature senescence in cotton in relation to potassium availability in soil: preliminary results. 10th Australian Cotton Conference. (Brisbane) pp 293-296.

Bell MJ, Harch GR, Bridge BJ (1995). Effects of continuous cultivation on Ferrosols in subtropical southeast Queensland. I. Site characterization, crop yields and soil chemical status. Australian Journal of Agricultural Research 46, 237-253.

Bell MJ, Moody PW (2001). Nitrogen, potassium, calcium and magnesium balances for dryland cropping systems (Burnett region, Queensland). In ‘Australian Agriculture Assessment Vol 1’. (National Land and Water Audit: Canberra, ACT).

Bell MJ, Moody PW, Harch GR, Compton B and Want PS (2008). Fate of potassium fertilizers applied to clay soils under rainfed grain cropping in northeast Australia. Australian Journal of Soil Research (In press).

Edwards NK (1998). Potassium. In ‘Soilguide: a handbook for understanding and managing agricultural soils bulletin 4343.’ (Ed. G Moore) pp. 176-180. (Agriculture Western Australia: Sth Perth).

Isbell RF (1996). ‘The Australian Soil Classification.’ (CSIRO Australia: Collingwood, Vic.).

Reuter D, Duncombe-Wall, D, Judson G (1997). Potassium balance in Australia's broadacre industries: a contemporary national and regional analysis. In: M Wong, Yash Pal and N Edwards (Eds.), Proceedings of the First Workshop on Potassium in Australian Agriculture, pp. 1-8. Geraldton, Western Australia.

White J (2002). Potassium distribution in ferrosols and its influence on rain-fed crop production in the South Burnett region of Queensland. PhD thesis, University of Queensland.

Previous PageTop Of Page