Evaluating the effect of crop rotations and tillage practices on soil water balances of selected soils and crop performances in the Western Cape

Swiegelaar, Nina Antionette (2014-04)

Thesis (MScAgric)--Stellenbosch University, 2014.

Thesis

ENGLISH ABSTRACT: The aim of this study was to investigate the influence of crop rotation and soil tillage on the soil water balance and water use efficiency of wheat, canola, lupin and medics in the Swartland sub region of the Western Cape. This trail was conducted as a component study within a long-term crop rotation/tillage trial during 2012 and 2013 at the Langgewens Research Farm (33016’42.33” S; 18042’11.62” E; 191m) of the Western Cape Department of Agriculture near Moorreesburg. The experiment was laid out as a randomized complete block, with a split-plot treatment design and replicated four times. Three crop rotation systems, continues wheat (WWWW), wheat/medic/wheat/medic (WMcWMc) and wheat/canola/wheat/lupin (WCWL) were allocated to main plots. . Each main plot was subdivided into four sub-plots allocated to four tillage treatments namely: zero-till (soil left undisturbed and planted with zero-till planter), no-till (soil left undisturbed until planting and then planted with a tined no-till planter), minimum-till (soil scarified March/April and then planted with a no-till planter) and conventional tillage (soil scarified late March/early April, then ploughed and planted with a no-till planter). All straw, chaff and stubble remained on the soil surface and no-grazing was allowed on all tillage treatments. Three replicates were included in this current study. Only the no-till (NT) and conventional till (CT) were included in this current study as main tillage treatments. The volumetric soil water content was monitored at weekly intervals during the active growing season (May-October) and once a month during the fallow period (November-April) using a Diviner 2000 soil moisture meter. The Diviner 2000 was used to record the soil water content at every 100 mm depth increment up to the maximum depth of the profile. At the end of the growing season the total biomass, grain yield and quality parameters were determined. The soil water balance data calculated from the 2012 season were found to be inconclusive due to too shallow installation of soil water monitoring tubes and big variations in the depth complicating any attempt in comparing data from treatments and cropping systems. Soil water monitoring tubes was installed to a depth of 900 mm in the 2013 season. Complications during planting in the 2013 season resulted in very poor emergence in the CT sites. Weed counts revealed that only 38 % of CT sites were covered by crop, 31 % with weeds and 31 % were completely bare. The NT sites had 40 % crop coverage, 50.5 % grass weed coverage and only 9.5% bare surface. As a consequences crop rotation had no effect on the soil water balance, while the tillage treatments showed a response. The effect that tillage had on the soil water balance was clearly shown in the 2013 season, in which 79 mm more rainfall occurred than the long-term average. NT retained more soil water in the profile in the drier first half of the season when only 30 % of the total rainfall in the 2013 season occurred. There was no real difference in the soil water retention in the second half of the season where 70 % of the total rainfall in the 2013 season occurred. Crop rotation did have a positive effect on grain yield. Wheat monoculture was out performed by legume based cropping systems. This trend was also observed in the biomass production. No significant difference between tillage treatments were recorded when comparing grain yield data. However wheat mono culture was again out-performed by the McWMcW, CWLW and LWCW systems producing on average significantly higher biomass. The data from both seasons suggest that in seasons where more rainfall than the long term average occurs, there is no difference in the RUE between cropping systems or tillage practices.. This study highlighted the major effect that the prevailing weather conditions have and that the expected advantages associated with NT most likely only come into play in dry conditions when plant water availability is limited.

AFRIKAANSE OPSOMMING: Die doel van hierdie studie was om die invloed van grondbewerking en gewasproduksiestelsels op die grondwaterbalans en doeltreffendheid van watergebruik te ondersoek in die koringproduserende gebied van Malmesbury. Hierdie eksperiment is uitgevoer as 'n komponentstudie binne 'n langtermyn grondbewerking/gewasrotasieproef gedurende 2012 en 2013 op die Langgewens Navorsingsplaas (33016'42 .33 'S; 18042'11 0,62' E, 191m) van die Wes-Kaapse Departement van Landbou naby Moorreesburg. Die eksperiment is uitgelê as 'n volledige ewekansige blok, met 'n gesplete perseel behandelingsontwerp met vier herhalings. Drie gewasproduksiestelsels naamlik, koring monokultuur (WWWW), koring/medic/koring/medic (WMcWMc) en koring/canola/ koring/lupiene (WKWL) is elk toegeken aan persele en vier keer herhaal. Elke hoofperseel is onderverdeel in vier subpersele en bewerkingsbehandelings is soos volg toegeken: Konvensionele bewerking (CT) - grond gebreek in Maart/April, en daarna geploeg en geplant met geen bewerkingsplanter. Minimum bewerking (MT) - grond gebreek in Maart/April en daarna geplant met 'n geen bewerkingsplanter. Geen bewerking (NT) - grond is heeltemal onversteur gelaat tot planttyd en daarna geplant met 'n geen bewerkingsplanter. Zero bewerking (ZT) - grond tot planttyd met rus gelaat en dan geplant met 'n sterwielplanter. Alle strooi, kaf en stoppels het op die grondoppervlak gebly en geen beweiding is toegelaat nie. Slegs drie herhalings is ingesluit in die huidige studie en slegs die geen bewerking (NT) en konvensionele bewerking (CT) is in die huidige studie as hoof bewerkingbehandelings ingesluit. Die volumetriese grondwaterinhoud is weekliks gemonitor tydens die aktiewe groeiseisoen (Mei - Oktober) en een keer 'n maand gedurende die braaktydperk (November - April) met behulp van 'n Diviner 2000 grondvogmeter. Die Diviner 2000 is gebruik om die grondwaterinhoud by elke 100 mm diepte tot die maksimum diepte van die profiel te bepaal. Aan die einde van die seisoen is die totale biomassa, graanopbrengs en kwaliteitparameters bepaal. Die data vir grondwaterbalans van die 2012-seisoen is buite rekening gelaat weens te vlak installering van moniteringsbuise en groot variasie in die dieptelesings wat enige poging om vergelykende data van rotasie en behandelings te verkry, bemoeilik het. Moniteringsbuise vir grondwater is geïnstalleer tot op 'n diepte van 900 mm in die 2013-seisoen. Komplikasies tydens die plantaksie in die 2013-seisoen het gelei tot 'n baie swak opkoms in die CT-persele. Slegs 38 % van die CT-persele was bedek deur die gewas en 31 % met onkruid, terwyl 31 % van die oppervlak onder CT-behandeling heeltemal kaal was. Die NT-persele het 40 % gewasbedekking, 50.5 % grasbedekking en slegs 9.5 % kaal oppervlak gehad. Dit het die poging, om die effek van wisselboustelsels op die grondwaterbalans, in die wiele gery. Alhoewel wisselbou skynbaar geen effek op die grondwaterbalans gehad het nie, het die tipe bewerking egter wel ‘n effek gehad. Die effek van grondbewerking op die grondwaterbalans het duidelik na vore gekom in die 2013-seisoen. In hierdie seisoen het 79 mm meer reën geval as die langtermyngemiddelde. Geen bewerking het meer grondwater in die droër eerste helfte van die seisoen in die profiel behou, toe slegs 30% van die totale reënval in die 2013 geval het. Daar was geen beduidende verskil in die grondwaterretensie in die tweede helfte van die seisoen toe 70% van die totale reënval in die 2013 geval het nie. Wisselbou het egter 'n positiewe uitwerking op die graanopbrengs gehad. Koring monokultuur is in opbrengsyfers geklop deur stelsels met peulplante as komponent. Hierdie tendens is ook waargeneem in die biomassaproduksie. Bewerkingsbehandelings het geen beduidende verskil in graanopbrengste tot gevolg gehad nie, hoewel die biomassaproduksie van koring monokultuur weer geklop is deur die McWMcW-, CWLW- en LWCW-stelsels. Die data van beide seisoene dui daarop dat in seisoene waar meer reën as die langtermyn gemiddelde voorkom, daar geen verskil in die RUE tussen verbouingstelsels of bewerkingspraktykes was nie. Hierdie studie beklemtoon die groot invloed wat die heersende klimaat speel en dat die verwagte voordele wat verband hou met NT waarskynlik slegs ‘n rol speel in droër jare.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/86237
This item appears in the following collections: