Rate and timing of nitrogen fertilisation for canola production in the Western Cape of South Africa

Coetzee, Albert (2017-03)

Thesis (MScAgric)--Stellenbosch University, 2017.

Thesis

ENGLISH ABSTRACT: Canola (Brassica napus) is increasing in popularity as a cereal crop in the Western Cape. Nitrogen (N) is generally the most limiting nutrient to canola production. Nitrogen fertiliser guidelines for canola are currently adopted from international literature or adopted from guidelines for wheat, and should be refined for the local environmental conditions. The aim of this study is to determine the optimal rate of N fertilisation, and the distribution thereof in the Western Cape. Field experiments were conducted in 2015 at Langgewens and Altona (Swartland) and Roodebloem (southern Cape). The trial was laid out as a factorial arranged in a randomised block design, with six N rates (0, 20, 60, 90, 120 and 150 kg N ha-1), of which each rate was applied in either one, two or three increments after planting, replicated in four blocks. Twenty kg N ha-1 was applied at planting. The rest was divided in equal increments either 30 days after planting (DAP), 30 and 60 DAP or 30, 60 and 90 DAP. Soil mineral N, leaf area index (LAI) and biomass was determined at 30, 60 and 90 DAP. Grain yield and thousand kernel mass (TKM) was recorded. Using these values obtained, agronomic N use efficiency (ANUE), water use efficiency (WUE) and profitability was determined by means of a sensitivity analysis. Treatments had no effect (P>0.05) on total soil mineral N content at any locality at any physiological stage with the exception of 90 DAP at Langgewens. Neither LAI nor biomass was affected by treatments (P>0.05) at any locality, at any physiological stage. Yield at Roodebloem was affected (P<0.05) by N fertilisation and treatments which had the highest yield were those who received 20 kg ha-1 at planting, and 70 or 100 kg ha-1 at 30 DAP respectively. These treatments did not differ (P>0.05) from treatments which received more than 60 kg N ha-1, regardless of the distribution. At Langgewens and Altona, N fertilisation had no effect (P>0.05) on yield, while TKM was not affected by treatments at any locality. No differences (P>0.05) were observed for ANUE at Altona and Langgewens, while treatments had a significant effect on ANUE at Roodebloem. The treatment that received 20 kg N ha-1 at planting and 100 kg N ha-1 at 30 days after planting and no N later, had the highest WUE, but did not differ (P>0.05) from a number of treatments that received more than 90 kg N ha-1 at various time intervals. The WUE at Altona and Langgewens was not affected by treatments (P>0.05). At both Altona and Langgewens the highest gross income was obtained by treatment that received no N at all, while at Roodebloem the highest gross income was obtained by applying 90 kg N ha-1 for the entire duration of the growing season. Preliminary results indicate optimum N fertiliser rate of 90 to 120 kg N ha-1, applied as 20 kg N ha-1 at planting and the remainder at 30 DAP at Roodebloem. No significant response to N applications was recorded in the Swartland. These results could be ascribed to drought conditions during 2015, which may have prohibited efficient uptake of N in the Swartland localities of Altona and Langgewens during critical periods. The differences between treatments at Roodebloem in the southern Cape might be due to a combination of sufficient rainfall over the growing season, lower temperatures and sufficient N being available during early, rapid vegetative growth. This in turn enabled a higher LAI which allowed for higher biomass accumulation, and consequently higher translocation to seeds. It is recommend that this study be repeated before results could be used to develop fertiliser guidelines for canola production in South Africa.

AFRIKAANSE OPSOMMING: Kanola (Brassica napus) se gewildheid as graangewas in die Wes-Kaap is aan die toeneem. Stikstof (N) is oor die algemeen die mees beperkendste voedingstof vir kanolaproduksie. Stikstofriglyne vir kanolaproduksie word huidiglik afgelei vanaf internasionale literatuur of die riglyne vir koringproduksie, en moet vir die plaaslike omgewingstoestande verfyn word. Die doel van hierdie studie is om die optimale N bemestingspeil, asook die verspreiding daarvan in die Wes-Kaap te bepaal. Veldeksperimente is in 2015 op Altona en Langgewens (onderskeidelik hoë en matige produksiepotensiaal areas in die Swartland) en Roodebloem (Suid-Kaap) uitgevoer. Die eksperiment is uitgelê as ʼn faktoriaal in ʼn ewekansige blokontwerp uitgelê met ses N-peile (0, 20, 60, 90, 120 en 150 kg N ha-1) wat in een, twee of drie stadiums na plant toegedien is, en in vier blokke herhaal. Twintig kg N ha-1 is toegedien tydens plant. Die res van die peil is dan verdeel in gelyke inkremente of óf 30 dae na plant (DNP), 30 en 60 DNP of 30,60 en 90 DNP. Grondminerale N, blaaroppervlakindeks (BOI) en biomassa is vasgestel op 30, 60 en 90 DNP. Graanopbrengs en duisend korrel massa (DKM) is na oes vasgestel. Hierdie waardes is gebruik om agronomiese N verbruiksdoeltreffendheid (ANVD), waterverbruiksdoeltreffendheid (WVD) en winsgewendheid deur middel van n sensitiwiteitsanalise, te bepaal. Behandelings het geen effek (P>0.05) op die totale grondminerale N op enige van die lokaliteite tydens enige van die fisiologiese stadiums gehad nie, met die uitsondering van 90 DNP op Langgewens. Behandelings het geen effek op BOI of biomassa (P>0.05) op enige lokaliteit, teen enige fisiologiese stadium gehad nie. Die opbrengs op Roodebloem is deur N behandelings beïnvloed, en die behandelings wat die hoogste opbrengs gehad het, het 20 kg N ha-1 tydens plant, en 70 of 100 kg ha-1 op 30 DNP onderskeidelik ontvang. Hierdie behandelings het nie verskil (P> 0.05) van behandelings wat meer as 60 kg N ha-1, ongeag die verspreiding daarvan, ontvang het nie. Op Langgewens en Altona het behandelings geen effek (P> 0.05) op die opbrengs gehad nie, terwyl DKM nie deur die behandelings op enige van die lokaliteite beïnvloed was nie. Geen verskille (P> 0.05) is waargeneem in ANVD op Altona en Langgewens nie, terwyl behandelings 'n beduidende effek op ANVD op Roodebloem gehad het. Die behandeling wat 20 kg N ha-1 tydens plant en 100 kg N ha-1 op 30 dae na plant en geen N later ontvang het nie, het die hoogste WVD gehad, maar het nie verskil (P> 0.05) van 'n aantal ander behandelings wat meer as 90 kg N ha-1 ontvang op verskillende tydsintervalle nie. Die WVD by Altona en Langgewens is nie deur behandelings (P> 0.05) beïnvloed nie. Op beide Altona en Langgewens is die hoogste bruto inkomste deur die behandeling wat geen N ontvang verkry, terwyl op Roodebloem die hoogste bruto inkomste verkry is deur 90 kg N ha-1 vir die volle duur van die groeiseisoen toe te dien. Voorlopige resultate dui optimale N-peile van 90 tot 120 kg N ha-1, toegedien as 20 kg N ha-1 tydens plant en die res op 30 DNP op Roodebloem. Geen beduidende effek op N bemesting is in die Swartland waargeneem nie. Die resultate in die Swartland kan aan droogtetoestande gedurende 2015 toegeskryf word, wat die doeltreffende opname van N tydens kritieke periodes kon verhoed het. Die verskille tussen behandelings by Roodebloem in die Suid-Kaap kan toegeskryf word aan 'n kombinasie van voldoende reënval oor die groeiseisoen, laer temperature en voldoende N wat beskikbaar was vroeg in die seisoen tydens vegetatiewe groei. Dit het op sy beurt die plante in staat gestel om 'n hoër BOI te ontwikkel, wat hoër biomassa akkumulasie, en gevolglik hoër translokasie na sade tot gevolg gehad het. Dit word aanbeveel dat hierdie studie herhaal word voordat dit gebruik word om bemestingsriglyne vir kanolaproduksie in Suid-Afrika te ontwikkel.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/100929
This item appears in the following collections: