Strategic nitrogen fertilisation of kikuyu-based pastures over-sown with legumes

Jordaan, Bernhard (2018-03)

Thesis (MScAgric)--Stellenbosch University, 2018.

Thesis

ENGLISH ABSTRACT: Kikuyu (Pennisetum clandestinum)-ryegrass (Lolium spp.) pastures for dairy production in the southern Cape of South Africa often receive nitrogen (N) fertilisation rates in excess of 500 kg ha-1 yr-1. To reduce the high N fertilisation inputs of the region, legumes have been incorporated into these grass pasture systems. Legumes fix atmospheric N and can improve nutritive value of the pasture. However, research concerning N fertilisation guidelines of grass-legume pastures in the southern Cape is limited. As result of this limited research, producers may be applying varying N rates to grass-legume pastures, which may result in these systems being environmentally and economically unsustainable. The aim of this study was to determine the effect of N fertilisation regimes on soil and pasture parameters of a kikuyu-perennial ryegrass (L. perenne)-clover (Trifolium spp.) (KPrCl) and a kikuyu-lucerne (Medicago sativa) (KL) pastures under grazing and irrigation. Five N fertilisation regimes were used as treatments and were applied after each grazing event [0 (N0), 20 (N20), 40 (N40), 60 kg N ha-1 (N60) and a variable N application (Nvar)]. Nitrate concentration measured from wetting front detectors as well the distribution of total soil N between the depths 0-100, 100-200 and 200-300 mm indicated that a significant amount of leaching may have occurred throughout the study. However, leaching was constant for all treatments on both pasture types. Furthermore N treatments had an minimal influence on total soil N, potential mineralisable N and urease activity. The only soil parameter that showed a significant reaction to N fertilisation regimes was total inorganic soil N. For both pasture types, total inorganic soil N increased under the N60 fertilisation regime. Seasonal herbage yield for the KPrCl type was only affected by N treatment in the spring (year one), where N60 produced a higher yield than N0 and Nvar. Treatment had no effect on total annual yield of KPrCl pastures. For the KL type, treatment N60 produced the highest or similar to the highest yield in the spring, summer and autumn. The annual yield under treatment N20, N40 and N60 were higher than N0 and Nvar. Botanical composition for both pasture types were affected differently by N treatments. For the KPrCl type, kikuyu was positively related to N fertilisation rates in autumn, whilst ryegrass was positively related to N rates in the winter and spring. For most seasons the clover component responded negatively to N fertilisation. For the KL pasture, kikuyu was not affected by N treatments. Lucerne only responded to N fertilisation in the summer, during which N40 had the highest lucerne component. The crude protein content of the KPrCl pasture type was highest or similar to the highest under treatment N60 for all months, whilst crude protein for the KL type was highest under N60 from February to June.

AFRKAANSE OPSOMMING: Dit is n algemene praktyk vir suiwelprodusente in die Suid-Kaap streek van Suid-Afrika om kikoejoe (Pennisetum clandestinum)-raaigras (Lolium spp.)-weidings teen stikstof (N) peile van meer as 500 kg N ha-1 jaar-1 te bemes. Om hierdie hoë N-peile te verlaag, word peuleplante saam met grasweidings gevestig. Peulplante het die vermoë om atmosferiese N na n plant beskikbarevorm om te skakel. Navorsingsinligting op N-bemsetingsriglyne van gras-peulplant weidings in die Suid-Kaap is egter beperk. As gevolg van die beperkte navorsingsinligting, word dit verwag dat produsente wisselende N-peile toedien. Dit kan lei na ekonomiese en omgewingsnagevolge. Daarom is die doel van hierdie studie om die effek van N-bememstingspeile op grond- en weidingsparameters van kikoejoe-meerjarige raaigras (L. perenne)-klawer (Trifolium spp.) (KPrCl) en kikoejoe-lusern (Medicago sativa) (KL)-weidings onder beweiding en besproeiing, te bepaal, Vyf bemestingspeile was getoets as behandelinge en was toegedien na beweiding [(N0), 20 (N20), 40 (N40), 60 kg N ha-1 (N60) en ’n wisselende N-peil (Nvar)]. Nitraatkonsentrasie wat vanaf benattingsfrontaanwysers bepaal is, asook die verspreiding van totale grond-N tussen dieptes 0- 100, 100-200 and 200-300 mm, het daarop gedui dat loging plaasgevind het, maar dat loging nie deur N-bemesting beïnvloed was nie. Totale grond-N, potensieël mineraliseerbare N en urease aktiwiteit was minimaal beïnvloed deur N-bemestingspeile. Die enigste grondparameter wat reaksie tot N bemesting getoon het, was totale anorganiese grond-N. Vir beide weidingstipes het totale anorganiese grond-N onder behandeling N60 deur die verloop van die studie verhoog. Seisoenale opbrengs van die KPrCl weidingstipe was net in die lente van jaar een deur N bemestingspeile beïnvloed waar die opbrengs van behandeling N60 hoër as N0 en Nvar was. Behandeling het geen effek op die jaarlikse opbrengs gehad nie. Vir die KL-weidingstipe was opbrengs in die lente, somer en herfs van behandeling N60 die hoogste of gelykstaande aan die hoogste. Die jaarlikse opbrengs vir N20, N40 en N60 was vir die eerste jaar van die studie hoër as N0 en Nvar. Botaniese samestelling van beide weidingstipes was deur N-bemesting beïnvloed. Vir die KPrCl-tipe was die kikoejoekomponent positief met N-bemesting in die herfs gekorreleer, terwyl raaigras positief met N bemesting tydens die winter en lente gekorreleerd was. Vir meeste van die seisoene was die klawerkomponent negatief met N bemesting gekorreleerd. Met betrekking tot die KL-tipe, het die kikoejoekomponent nie op N-peile nie gereageer. Die lusernkomponent het net in die somer op N bemesting gereageer. In die somer was die lusernkomponent van N40 hoër as die ander behandelings. Die ru-proteininhoud van die KPrCl-tipe was onder behandeling N60 die hoogste of naaste aan die hoogste, terwyl ru-proteininhoud van die KL-tipe die hoogste vir N60 vanaf Februarie tot Junie was.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/103384
This item appears in the following collections: