Conservation tillage research and development in South Africa by Richard Fowler ARC-Grain Crops
Institute P/Bag X 9059, Abstract South Africa has a wide range of soils and climates. Extensive studies, especially over the past 25 years, have been conducted into practices and systems which could conserve the soil, water, energy and other resources required for sustainable crop production under various of these soil and climate combinations. Some of the results obtained and ongoing or envisaged work is outlined. It is hypothesised that, given the co-operation and support of international agencies and in co-operation with other African countries, the South African research system, spearheaded by the National Department of Agriculture and its Agricultural Research Council, could play a major role in achieving sustainable national and household food security in the sub-continent. 1. Introduction South Africa has a total area of 122 341 000 hectares, of which only one third is level to moderately sloping and only 14% is arable. Appropriately 1.2 million hectares is irrigated, accounting for almost 30% of crop production and 4% of the Gross Domestic Product (Scotney & van der Merwe, 1992; Goodland, 1995). The cold Benguella (Atlantic ocean) current running up its west coast and the warm Mozambique (Indian Ocean) current running down its east coast account in part for the climatic extremes in South Africa – ranging from semi-desert in the west to semi-tropical in the east. The Southern tip, including most of the deciduous fruit and some wheat producing areas, receives most of its rain in winter. In the remainder of the country, erratic, spaced storms of high intensity, normally commence in September-October and continue, apart from a frequent mid-season break in January-February, to March-April. The mean national annual rainfall is appropriately 450 mm, with 75% of the country receiving less than 600 mm. Only 3% of the arable area of the country is classified as being of high agricultural potential, almost 60% of the soils having a very low organic matter content conducive to land degradation and low productivity (van der Merwe & de Villiers, 1998). Maize is both South Africa’s staple food and most extensively (1.7 – 4 million hectares) grown field crop, followed by sugarcane and wheat. Maize in particular is often grown in areas of marginal rainfall or soil depth, together with some sorghum and cotton. In the past a net exporter of food, increasing population densities and past mismanagement of arable and pasture lands are threatening South Africa’s natural resource base and national and household food security. Major causes of soil degradation are organic matter depletion and soil sterilisation. These in turn cause acidification or alkalinasation, compaction or crusting, runoff, erosion, infertility and desertification (van der Merwe & de Villiers, 1998). Conservation tillage has the potential to arrest or reverse these processes, and its wide scale advocacy and adoption is therefore of national importance. 2. Background and conceptual basis 2.1 Conservation tillage pre-1950 Dominated by Euro-American thinking and the drive to "modernise" by maximising production utilising increasing quantities and types of external inputs, South African researchers and extensionists have only recently attempted to discover and understand indigenous conservation principles and practices (Oettle et al. 1998). An example of these is the terracing used by Venda farmers in the Northern Province (Critchley & Netshikovehla, 1997), but there are undoubtedly many more. In the main, however, South African tillage principles and practice this century (except possibly in the past 5 years) have been almost exclusively derived from those expounded or utilised in Europe and North America. Thus Leppan & Bosman (1923), although they define tillage as "the manipulation of the soil by means of implements so that its structural relationships may be improved for crop growth", state that two of the objectives of tillage are ‘to pulverise the soil’ and ‘to place beneath the surface manure, stubble, stalks and other organic matter, where it will be out of the way’. The most important function of tillage was to prepare a good seedbed, inter alia ‘free from weeds and surface trash’. Tillage implements fall into two categories – those that loosen the soil e.g. ploughs, and those that compact it e.g. rollers. The primary functions of the plough were to shear off, split the furrow slice vertically and horizontally, resulting in the complete pulverisation of the soil, and ‘put any rubbish beneath the surface’. Of the three types of plough available, the disc plough had a slightly lighter draft and was especially effective on very hard dry soil and for covering trush. The Lister Plough was a double mouldboard which left the ground with ridges. Even then, however, it was recognised that moisture was the chief factor limiting crop production in South Africa, and winter ploughing was advocated so that the soil would face winter in a ‘rough condition’, conducive to weathering and receptive to the spring rains. Soils would then be ‘pulverised’ into a fine tilth, although the efficacy of preventing evaporation by dust mulches was much disputed. Mundy (1923) advocated working land in both directions. Ploughing, he believed, was the most important tillage operation, although there was little African evidence in favour of sub-soiling or deep ploughing. The low maize yields obtained were often due to the postponement of autumn ploughing, To counter erosion new land would be on as gentle a slope as possible, and ploughed and planted on the contour. If the size of land permitted, ‘a useful practice was to leave natural roadways of unploughed veld parallel to the slope’. Saunders (1930) believed the primary objective of tillage was to make the soil ‘a place of hold-fast or anchorage for the plant’, although other objectives included the soil incorporation of surface organic materials so they could be readily converted into humus. The ox was still the main source of tradition in South Africa. The main tillage operation was ploughing, but large heavy tine cultivators had recently been introduced ‘to take the place of the plough in certain cases’. On the heavy black ‘turf’ soil of the Springbok Flats some farmers were ploughing only once every 2-3 years, using tines in the other seasons. The cultivator loosened the soil sufficiently but ‘the only effective method of turning under organic matter was ploughing’. Wind erosion, especially on the sandy soils of the (now) Free State, could be reduced by not winter ploughing, winter ploughing but leaving in a very rough state, or using the Lister plough to throw up ridges. Trials at Pietersburg (1923-28; Nothern Province) and Glen (1922-27; Free State) on red sandy and dark clay loams had indicated increasing yield advantages from 7 inch (180mm) or 10 inch (250mm) ploughing compared to 4 inch (100mm). This, it was felt, was due to the ‘burning out’ of shallowly incorporated organic matter. Soil mulching required further investigation. Thompson (1936) concluded that, the creation of a soil mulch had doubtful value in South Africa. Evaporation appeared to be the principal dissipating factor of rainfall, and was minimised by the presence of a dense soil cover. Van Reenen (1935) found the most prolific cause of water and soil loss on ploughed land was ploughing up and down the slope. Ploughing should be done along the contours and, in (now) KwaZulu-Natal, the use of contour strips of grass had been found effective. In 1933 the Government had agreed to assist farmers check erosion, and local soil conservation committees were being established. The first half of this century was undoubtedly a difficult period for all farmers, the rapidly increasing demand for agricultural products at often uneconomic prices leading to a systematic overtaxing of both grazing and arable lands. The silt load of some of the country’s principle rivers during the 1919/20 season was estimated to be 187 million tonnes, which grew in the 1950’s to an estimated average per annum soil loss of 300 million tonnes (Ross, 1957). Soil depletion became more and more pronounced, giving rise to widespread erosion and desiccation which the Soil Conservation Act (No.45 of 1946) was introduced to help, through subsidy and regulation, to curb. A new era of Conservation Farming was borne. 2.2 Conservation farming in South Africa post-1950 Klintworth (1957) identified the three most important factors contributing to the physical decline of soils as surface pulverisation raindrops; solar desiccation and heating; and trampling by stock (Figure 1). All these could be reduced by the maintenance of a mulch of straw or semi-decomposed compost on the soil surface, which would also encourage the build-up of soil organic matter and the penetration of soil roots. Early work on conservation tillage in maize production involved stubble mulching and reduced tillage with tine of ‘chisel’ ploughs. Maize tillage research at Potchefstroom (now North West Province) over the 9 seasons 1955/56 – 1963/64 showed that reduced tillage had little or no adverse affect on yields. On a sandy soil at Viljoenskroon (now in North West Province) ripping to a depth of 450mm under the row produced significantly higher maize yields during the low rainfall years 1978/79 and 1979/80, while mouldboard plough treatments caused compacted layers below their working depth (Mallett, Koch, Visser & Botha, 1985). Owens, 1984 reported that sandy soils were particularly susceptible to compaction, as were heavier soils if worked wet or with heavy equipment. 2.3 Wind erosion Approximately 2.5 million hectares in South Africa is prone to wind erosion. This is observed mainly by maize and wheat producers on sandy soils in the North West and Free State Provinces. Wind may remove soil from bare areas, and may in the process damage young plants. At a wind erosion symposium in 1974, four measures for the control of wind erosion were identified, namely:
2.4 Soil-water and tillage interaction The most important elements of weather, which influence crop performance, are radiant energy and moisture (Mallet & de Jager, 1974). The amount of either which reach or remain in a soil can be influenced by the vegetative cover and especially the crop residues on the surface of the soil, which in turn can be influenced by the type and intensity of tillage. Thompson (1965) working with sugarcane in (now) KwaZulu-Natal showed that the trash blankets improved soil moisture levels which resulted in increased yields. Mallett & Johnston (1983) compared the effect of conventional tillage and direct drilling on maize yield and the physical characteristics of a Doveton series soil (30% clay). After 8 years continuos maize they found direct drilled plots showed slightly higher compaction in the top 120 mm, but below this conventional tilled plots were more compacted. Organic carbon was higher in the surface layers of the direct drilled plots. Soil moisture in the direct drilled plots was much higher at planting, resulting in significantly higher yields in seasons of below-average rainfall. Berry, Mallett & Johnston (1985) confirmed these findings in experiments incorporating four tillage treatments,with % maize stover cover at planting (shown in brackets):
These trials showed that tillage practises which maintain higher levels of surface residues retained more water. Figure 1: The consequences of inappropriate land use (Klintworth, 1957).
At the commencement of the second season of one of these experiments the plant available water (PAW) in the top 600mm of the soil profile of each tillage treatment was not significantly different. However, following primary tillage, PAW decreased with increasing tillage, attributed to the reduced evaporation prior to planting under the increased soil cover and in the less disturbed soil. Maximum soil temperatures at 50mm increased with decreasing residue, resulting in quicker seedling emergence and leaf area development during the early vegetative growth (Berry et al., 1987). 2.5 Surface residues and soil loss Land & Mallett (1982) investigated the effects of various methods of primary tillage and seedbed preparation on surface residues. They found flail equipment to be more satisfactory than discs to cut up maize stover, as when the discs were set sufficiently aggressively they buried too much stover. The chisel plough left the most surface residue, twisted shanks being less affected by speed of operation but burying more trash than straight shanks. A swanson-type rainfall simulator at Cedara in KwaZulu-Natal was and is still being extensively used to compare the effects of simulated rainfall events on soil and water loss from soils under different soil covers and subjected to different tillage treatments. Land & Mallett (1984) compared the effect of six levels of maize stover providing 0-75% soil cover on a clay loam with 3.5% slope. Increasing percentage cover increased infiltration and reduced soil loss, with a minimum 30% cover required to keep both within acceptable limits. A number of natural runoff plots were set up at Cedara in mid-1982. Soil type was a 2 metre deep Inanda clay (48% clay), slope 6.9 - 8.8%. Acceptable soil loss was 6.75 t/hectare (Table 1). 2.6 Weed control Numerous workers have noted the tendency for weed spectrum to change under reduced tillage. Weed control in the long term tillage trials at Cedara became a major constraint, and it inspired investigations by P.E.L. Thomas. This study produced a system of control. In brief, the slower emergence and initial growth of direct drilled maize occasioned by the lower soil temperatures under mulch early in the season, together with the interception of pre-emergent herbicides by the mulch, allowed weeds such as Digitaria sanguinalis and Cyperus esculentus to successfully emerge over a longer period. Higher doses of the pre-emergent herbicides or slow releasing formulae (such as some granules) were of assistance. Direct post-emergent sprays of what came to be termed the ‘Cedara Blitz’ (paraquat mixed with a triazine with or without a top-up of the pre-emergent in high potential maize) provided the final solution. The application of post crop-senescence herbicides to reduce weed seed set and the control of winter weeds to conserve soil moisture were also studied. 2.7 Long-term effects A severe rainstorm in February 1985 caused severe lodging in all the maize on Cedara except that planted by no-till. Investigation showed this was due to the firm anchorage of the no-till plants (Lang, Mallett & Berry, 1986). After 8 years, direct drill maize, on a Hutton/Doveton clay loam, it was found that the top 120 mm had become denser, and that organic carbon levels in the top 20mm was higher than in conventionally tilled plots. Four years later it was found that densities had stabilised but organic carbon levels had increased from 3.8 – 4.7%, compared to 3.3% in the conventionally tilled plots. Surface pH levels were lower and P, K and Al levels were higher in the direct drilled plots. Earthworm counts were significantly higher (Mallett, Lang & Arathoon, 1987). Agenbach & Maree (1989), working in a winter rainfall area on a shallow sandy-loam soil with a gravel and stone content of 45%, compared the effects of no-till, tine and mouldboard based tillage systems in wheat monoculture and wheat-after-legume pasture systems. Most soil parameters were improved by less intensive tillage. Lawrance et al. (1998) analysed the long-term tillage trial at Cedara. Over the 13 seasons, the average yields from the reduced tillage treatments differed by only 10 kg/hectare, but exceeded the conventional tillage by 100 kg/hectare. Table 1: Average annual soil loss (adjusted to a 9% slope) and grain yield from the Cedara runoff plots for the period 1983, and soil organic matter percentage in the top 150mm after ten years Continuous maize (after Russell & Gibbs, 1995).
Production costs were lowest for the direct drill treatment due to higher yields as well as fuel, labour and machinery savings. Other factors favouring reduced tillage systems were reduced soil erosion, increased moisture retention and the ability to plant earlier. 2.8 Conservation tillage research 1988-1998 The ARC-Grain Crops Institute Tillage Trial reported above was split to include a lime and a paraplough treatment in 1988. The acidification of the surface in especially no-till was attributed to excessive nitrogen applications, especially in drier years, and some subsoil compaction was apparent (Table 2). Subsequently the importance of crop rotation was noted and the trial again split to incorporate soya beans. The growing awareness in South Africa of the importance of animal traction led Fowler (1996) to split another similar tillage trial two ways. The trial had been established in 1983 and incorporated No-Till, Chisel, Chisel & Disc and Mouldboard & Disc (conventional) tillage treatments. In 1994 the trial was split both ways, all the maize crop residues being removed from two sub-plots and the others cultivated with similar equipment drawn by a 70kW John Deere tractor or 4-6 oxen. The mean yields from the two traction methods were identical, but the removal of the crop residue 5 months prior to planting reduced yield and plant density. Unfortunately the trial site had then to be vacated. Other workers however, from IMAG-DLO of Netherlands, ARC-IAE at Silverton and Fort Hare University, are pursuing animal traction based conservation tillage research. On-farm conservation tillage trials based on tractor, animal draft and manual means are being conducted in the KwaZulu-Natal Department of Agriculture. The writer is working with them in a Monsanto funded project. No-Till crops like maize, cotton, and possible rotation crops such as soyas, cowpeas and dry beans are being tried. It is hoped some 200 on-farm demonstrations will be co-operatively established. 2.9 Conservation tillage adoption and research needs One major constraint to the adoption of conservation tillage is the possibility of pest and disease surviving over-winter on the crop residues, especially if no crop rotation is practised. This led to an extreme reluctance on the part of wheat farmers to adopt conservation tillage practises and two major reversals in conservation tillage adoption by maize farmers due to diplodia cob rot and grey leaf spot (GLS) outbreaks. Tolerant cultivars to both these diseases are, however, available. Table 2: The effect of lime and paraplough treatments on the ARC-Grain Crops Institute Tillage trial (after Berry, 1998).
Other constraints include deficiencies or gaps in the information available to conservation tillage farmers. This occurs due to poor communication between farmers, advisers, researchers or because work under South African conditions is still required. Notable deficiencies include:
Additional practitioner queries are presented in Table 3. Table 3: Queries listed by the No Till Club, KwaZulu-Natal, South Africa in an adviser-defined order of priority and circulated to members for their comment in September 1998. _____________________________________
____________________________________ Berry (1998) emphasised the unsuitability of no-till for sandy soils (Table 4). Partly due to the constraints outlined above, but also due to the conservatism of advisers and practitioners, no-till has had extremely limited acceptance. Reduced tillage is however practised by many large scale commercial farmers, especially those cultivating sandy soils. No-till for vegetables and cotton producers is virtually nil, with the only adopters of any significance being some sugar farmers in KwaZulu-Natal. This is the case for wheat, medic, lucerne and canola farmers in the Western Cape, and maize, wheat, soya farmers in KwaZulu-Natal. In both these cases, only 2-3% of the area is under no-till. Table 4: Probable maximum number of seasons of no-tillage intervention required (after Berry, 1998).
Conclusion Considerable research and development of conservation tillage techniques has been conducted in South Africa, especially in the past 25 years. Much of this knowledge has still to be effectively digested and presented to potential practitioners, especially small scale farmers. Much work, especially on animal traction, crop rotation and acidification, remains to be done. The reduced research capacity in notably the Agricultural Research Council (ARC), combined with the lack of motivation of many workers to engage in non-income generating research, or to recognise or try to understand the unique problems inherent in small scale farmer systems, may severely limit the investigative capacity of the South African system in the future. However, the prioritization by the South African Government of National Unity and its National Department of Agriculture of the protection and proper utilisation of the country’s natural resources, and its expressed desire to be of service and assistance to its neighbouring countries and own resource poor farmers will help. This, combined with the enthusiasm and commitment of the newly formed No-Till Club in Kwa Zulu-Natal will make the difference. The club has senior officials and researchers in the KwaZulu-Natal Department of Agriculture; Vice-President Frans Hugo and various Directors and researchers in the ARC; and other individuals in the universities and other departments in South Africa. They will ensure that conservation tillage receives the support it needs. This support is effectively linked to international experience and funding and is backed by the information and the requirements of our fellow Africans. The research and development capacities of South Africa will therefore play a major role in the rapid sustainable adoption of appropriate conservation tillage practices and systems in the sub-continent. Such adoption will have a major impact on ensuring national and household food security. It could also stabilise the natural resource base of the region and reduce production and living costs. This will lead to reduced unemployment and conflict. It will in turn facilitate the African Renaissance and Peace. References
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Table of Contents