Understanding 'Challenger' sweet corn yield, quality and phenology responses to phosphorus

Abstract

The research in this thesis examined the response of field grown 'Challenger' sweet corn crops to P supply at Lincoln Canterbury, New Zealand in 2001/02 and 2002/03. Initial soil tests showed that the site had an available soil P (Olsen, bicarbonate extraction) of 6 µg ml⁻¹. In both 2001/02 (0, 50, 100, 150 and 200 kg P ha⁻¹) and 2002/03 (0, 50, 110, 170 and 240 kg P ha⁻¹) five rates of P fertiliser were applied to these crops. The kernel yield and biomass responses to P were then examined. Initially, a traditional empirical approach was used to analyse the yield responses to a P fertiliser. This described the asymptotic increase of both kernel yield and crop biomass to increasing P supply kernel yield ranged from 2.0 to 4.4 t DM ha⁻¹ in response to P fertiliser. These P responses were specific to this experiment. Therefore to provide greater understanding and insight into crop growth the mechanisms of these responses were examined in detail. An adjacent experiment with 5 rates of N fertiliser (0, 45, 90, 180 and 300 kg N ha⁻¹) showed that N had only a minor impact on kernel yield of sweet com and therefore these treatments were not studied further. Except that these data were included in a preliminary analysis of DM partitioning in response to crop DM produced.

The increased kernel yield with P fertiliser was associated with changes in total crop biomass (9.7-15.7 t ha⁻¹). However, the partitioning of this biomass was conservative with 24% kernels, 44% stems and leaves and 32% ears. Ear quality, unfilled tip length and individual kernel mass, was also from crops with the greatest biomass. The application of P fertiliser also decreased the time from crop emergence to canning maturity by 6-7 days. This acceleration was caused by an 80-115 °Cd (Tb = 8°C) decrease in the period from crop emergence to silking.

The causes of the differences in crop biomass were then investigated in terms of radiation interception and use. Total accumulated intercepted solar radiation (RIcum) was 23 and 39% greater when 200 or 240 kg P ha⁻¹ was applied in 2001/02 and 2002/03, respectively, compared with the 0 kg P ha⁻¹ crops. This was due to both a faster leaf appearance rate and a greater area of each individual leaf. The phyllochron (ºCd between successive leaf tips) was 4-6 °Cd longer in the 0 kg P ha⁻¹ than in the 200 and 240 kg P ha⁻¹ crops in both seasons. The appearance of fully expanded leaves showed a similar pattern with a delay in the 0 kg P ha⁻¹ crops compared with the crops receiving P fertiliser. The area of individual leaves followed a bell shaped curve. The largest leaf was consistently leaf 11 or 12, which was 29-37% larger when P was applied, compared with the 0 kg P ha⁻¹ crops. The addition of P fertiliser had only a minor effect on the final number of mainstem leaves (16.7-18.0). P fertiliser did not affect either the rate at which the fraction of senesced leaf area increased or the extinction coefficient (0.65).

The radiation use efficiency (RUE) was consistent with a previously established temperature response and unaffected by P supply. However, during the early phases of crop growth (RIcum<134 MJ m⁻²) RUE was only -50% of the 1.3 g DM MJ⁻¹ found for the majority of the crop duration. The mechanisms responsible for this were unclear and require further examination. With the exception of the first and smallest leaves (<8), the specific leaf phosphorus (SLP) was always greater than 0.1 g P m⁻² and photosynthesis was unaffected by P supply. In the first leaves the SLP was less than 0.1 g P m⁻² for the 0 kg P ha⁻¹ plots and consequently there was a minor decrease in leaf photosynthesis in these crops during this phase.

The response to P supply of crop maturity, phyllochron, fully expanded leaf appearance, individual leaf area and RUE were incorporated into a simple framework for simulating yields. Combining this with long term weather data from both Hawkes Bay (1976-2002) and Lincoln (1960-2003) showed that a single yield response curve would have been inappropriate for multiple sites and seasons. Although P fertiliser had a marked effect on simulated kernel yield, the greatest source of variability in simulated yields was from seasonal weather variations. Insufficient P also increased the risk of a crop failing to mature before the first autumn frost by -10% at Lincoln from a range of typical sowing dates. In contrast this risk was minor at Hawkes Bay.

Overall, the dominant effect of P supply on kernel yield in 'Challenger' sweet com was on RIcum, with conservative RUE and DM partitioning. Further research should aim to isolate the mechanisms by which P supply limited individual leaf areas and leaf appearance rates. These data could be linked to a mechanistic model of soil P uptake to form a powerful research tool for analysing P responses for crops in other sites and seasons.