Herbicide Efficacy under Semi-controlled Conditions

Oneof the main causes of poorweed control on sugarcane farms is the occurrence of weed germination in periods during which demands on the farm's resources are at a maximum and during which wetconditions may prevent timely herbicide application or other weed control operations. It would therefore facilitate weed control if weed germination could be predicted from meteorological data and a knowledge of weed germination patterns. For this reason two experiments were conducted in pots under semi-controlled conditions: the first to study the germination patterns of 4 grass weeds in 5 different topsoil horizons under two different irrigation regimes; and the second to investigate the effects ofalachlor on 5weed species underdiffering moisture conditions. The results indicated that weed germination patternsvariedwithsoiltypeand moisture conditions. Survival of weeds was lowest in soils with low water-holding capacities under rainfed conditions. In spite of slower weed germinationin heaviersoils a greater numberofweeds survived in theexperiment. Theefficacy ofalachlor depended on rates of application which appeared to be more important than the timing of water application. Introduction Weed control in sugarcane generally entails using an integrated programme involving herbicides, hand labour, and in some instances mechanical cultivation. Programmes are designed to fit in with seasonal requirements; ie in winter, when few grass weeds occur, growth of broadleafweeds is slow and herbicides are less effective, hand labour is more cost-effective; while in spring and summerundermoistconditions growth of grasses, broadleafweeds and watergrass is vigorous, and herbicides are more effective. In practice, all fields harvested duringtheearlypartoftheharvesting season will require some weed control measures in spring when conditions become suitable for weed growth, as it is unlikely that the crop will have formed an adequate leafcanopy by then. This places a demand on labour at a time when both planting and fertilizing operations are being carried out. The use of herbicides in spring and summer has been suggested by Lonsdale,' Rau,' and Wise" as a means of alleviating this problem. Each of these authors emphasised the needfor programme planning and careful timingof herbicide applications to achieve acceptable weed control throughout the season. In practice, in spiteof a wide choice of herbicides, weed competition still occurs often in sugarcane fields. Although programme planning helps considerably in making contingency plans, the most difficult aspect ofweed control is probably deciding when to begin herbicide treatments in spring and when to change from short term treatments to long term residual treatments (which rely on adequate moisture conditions). For this reason it would be useful to define the relationships between the environment and weed germination patterns, as well as the conditions under which herbicides act most favourably. In particular it would be useful to be ableto predictweed germination in various soils using meteorological data, and to predict the performance of a particular herbicide under specific meteorological conditions. A great dealof workhas been carried out under controlled conditions to relate various environmental factors to herbicide efficacy. However Gerber et al.' indicated that laboratory experiments have tended to be focused on single factors, and conclusions can be misleading when applied to field conditions. It was for this reason that the twoexperiments reported herewere conductedunder conditions closely resembling those occurring in the field, so that particular factors could be examined without eliminating theeffects ofotherenvironmental factors. Relationships between soil characteristics and herbicide performance have been studied extensively, but less work has beenconducted on the effects of soil type on weed germination. The objectives of the two experiments reported here were to observe grass weed germination patterns in different topsoils during September, October and November in relation to soil moisture and temperature, and to examine the effects of moisture regimes on the efficacy of alachlor, a commonly used pre-emergence grass-killer. Topsoil ppm Soil Form Series Horizon N cat" pH P K S Ca Mg Na Zn Al Fe I Hutton Clansthal Orthic 2 8,70 80 35 83 1800 34 21 1,3 0 12 2 Fernwood Fernwood Orthic I 4,80 28 65 27 112 28 0 1,0 5 113 3 Kroonstad Kroonstad Orthic I 4,85 24 87 39 280 86 0 1,5 8 224 4 Arcadia Rydalvale Vertic 3 5,45 30 319 47 1565 220 0 4,0 0 66 5 Milkwood Milkwood Melanic 2 8,35 35 175 162 1800 184 41 2,1 0 30 Table 1 Results of analyses of 5 topsoil horizons • N Cat = nitrogen mineralisingcategory I = very low 2 = low 3 = medium