The tropical Zero Tillage (ZT) technology was forged by settlers from sub-tropical South Brazil and regional farmers with knowledge of ZT, aided initially by herbicide manufacturers and several separate private-sector supported R & D efforts. Introduction of ZT in tropical Brazil coincided with a significant elevation of technology in arable crops and this technology promotes the use of rotations and integration with livestock operations. The Ässociação de Plantio Direto no Cerrado" (APDC) has become a focal point in disseminating ZT technology, promoting Friends of the Soil Clubs (FSC’s), collaborating with government agencies and obtaining policy changes. Planters and drills, for mechanized and animal traction, are made in Brazil and the introduction of double disc openers, with or without differential diameters and Brazilian patent fertilizer knife/guillotine disc cleaner markedly improved planter r. Emphasis is placed on second cropping and crop rotation for adequate crop residue production. Millet (P. typhoides) was introduced and is widely used pre-plant, post harvest or with aerial oversowing on soya shortly before or after leaf fall, as a mulch generator, leading to possible reductions in herbicide use. The pre-plant dessicant herbicides are neutral to the environment, while the crop residue and soil organic matter increases act as a vast carbon sink. The system dramatically reduces erosion losses when compared to Conventional Tillage (CT) raises farmers’ technical management capacity, generates a high degree of social responsibility and produces many positive spinoffs to the environment, at little or no cost to society. 

Introduction 

Brazil’s experience, with nearly 3 million hectares of, Zero Tillage (ZT) in tropical arable crops, from the tropic of cancer to 3º North, shows that ZT farming offers the only long term solution for sustainable agriculture in the tropics. This paper will demonstrate why. The next step is to ensure that this technology development and transfer experience will be utilized in other countries. 

We have adopted the term „Zero Tillage" here to portray the radical 180º turn in thinking which the technique engenders. Our experience is that use of existing terminology, not specific to Zero Tillage (ZT), does not assist the wholehearted adoption of the new system, which is fundamental to success. The same is true for "green manure", which implies the now-superseded concept of ploughing green manure down, whereas in Zero Tillage, the place for crop residues is on the surface; thus we use the term "cover crop". 

Historical Background 

Manual Zero Tillage methods, with some form of hoe or digging stick, have been in regular use by indigenous peoples and other small farmers in the Americas for centuries, if not millenia. The „tapado" system of cutting fallow to cover bean seed in Central America and Northeast Brazil exemplifies this. However, in Latin America, the first recorded attempts at mechanized Zero Tillage began in sub-tropical Brazil, with research tests in 1969-1972 in Rio Grande do Sul and Paraná states (humid sub-tropics) and at farm level in1972 (Borges, 1993).Paraquat was the principal dessicant herbicide in use and there were very few post-emergent options. Pioneer farmers spurred the development of this technology in both the sub-tropics and the tropics and the state research institute of Paraná gave support to the early development in that state (IAPAR. 1981) ; federal research was slow to respond. Figure 1 shows that early growth in Brazil was extremely gradual, due to planter and herbicide limitations and significant incremental costs of adoption estimated at up to of direct costs (IAPAR, 1981). Development in the tropical region was also slow at first but the rapid expansion phase occurred with a shorter lag on the first attempts, starting with better planter and herbicide choices. Farmer NGO’s, and private sector support were a fundamental element in this development. 

In the tropics, from about 1980 onwards, small farmers in Rondônia state were using paraquat to control weeds in rice straw before zero tilling beans, as a second crop, with a commercial manual jab planter. First records of mechanized Zero Tillage in the tropics of Brazil, and probably worldwide, were in the state of Goiás in the central elevated wet/dry savannah region („Cerrado"), dating from 1981/2, at farm level (Landers, 1994). There was an inter-regional technology transfer of the basic principles of ZT through migrating farmers, from sub-tropical to tropical Brazil, but the agronomics of a dry tropical winter versus a wet sub-tropical winter with frost had to be worked out. 

^{1/Founder member and Executive Secretary of the Ässociação de Plantio Direto no Cerrado (APDC)}

Before Zero Tillage Reached the Cerrado Region 

In 1980 there was no Zero Tillage in tropical Brazil, where modern agriculture has developed principally in the 200 million hectare wet-dry savannah region of Brazil’s central plateau, the „Cerrado". The soybean area in this region increased by over 2,5 million hectares between 1980 and 1988 and now exceeds 4 million hectares. A large proportion of monoculture on highly limed/fertilised tropical latosols and quartz sands, under offset disc cultivation for weed control and fine seedbed preparation (to ensure maximum effectiveness of the predominant pre-emergent herbicide, Trifluralin), pulverized the soil while the meagre soybean residues with a high C:N ratio meant that organic matter levels dropped rapidly, with consequent loss of soil aggregates (Resck et al., 1991 and Silva et al., 1994). Disc pans, surface capping and clay eluviation reduced rainfall infiltration in a region where rains of over 100mm in a few hours are not infrequent. This resulted in considerable erosion losses between contour terraces and overrunning of the terraces, with serious gullying, up to 10% replanted area and difficulty in meeting planting schedules. Annual soil losses were of the order of 2-20 ton/ha, varying chiefly with slope and rainfall intensity (Chaves, H.M.L, personal communication) while the same author (Chaves, 1997) indicates that soil losses under CT exceed the regeneration rate of the soil and increase the need for agricultural inputs. 

Technology Development and Transfer of Tropical Zero Tillage in Brazil. 

In 1985/89 total ZT area was about 35,000 ha; by 1992, this area had risen to an estimated 180,000 hectares. With the activities of pioneer farmers, several private sector sponsored on-farm research projects (Cooperlucas/RPA/CRAD in Mato Grosso, Fundação MS and Sementes Bonamigo and Fundação MS, in Mato Grosso do Sul and Manah/JL Associates in Goiás). With the exception of the latter, these are still ongoing and sevearla new farmer R&D foundations have been founded, including the states of Maranhão, Bahia and the Federal District. The CIRAD-led initiative has expanded to cover ZT in cotton in São Paulo and Goiás (RPA/Grupo Maeda/CIRAD). With these results, new developments in ZT planters and results of official and private sector herbicide research, enough information, specific to ZT, had been generated to promote the practice generally. Official research, after an initial lag, is beginning to bring forward results and efforts are expanding to prepare for second-generation developments in this new system (Ayarza et al.1997, Spehar and Landers 1997). The new dynamics of soils, pests and diseases and introduction of new crop species are of high priority. 

With the sole objective of promoting ZT, the NGO „Associação de Plantio Direto no Cerrado"¹ (APDC) was formed in 1992. In collaboration with input and machinery manufacturers and suppliers, APDC began organizing short courses field days and seminars, culminating in the National Zero Tillage Meetings in 1996, with 2300 participants, a 30 hectare demo area with 30 commercial firms exhibiting. This will be repeated in Brasília in June 1998, with a forecast 3000 participants. A private sector support group for ZT was founded in 1992 of herbicide, seeds and fertilizer firms and is still actively supporting these activities, many other agribusiness companies participate on a case by case basis. 

Local, municipal-level „Friends of the Soil" clubs (FSC’s) have been organized and these use farmer-to-farmer communication, swapping experiences and encouraging new Brazilian network of small private technical assistance offices and technical departments of co-operatives have also been active in promoting ZT. In 1993, the APDC produced a de facto field manual, based largely on farmer practice and official/commercial herbicide recommendations plus some common-sense extrapolation of principles from the sub-tropical South of Brazil (4300 copies). In 1996 a quarterly technical bulletin was inaugurated (circulation 6000) and in 1997 a book entitled „Zero Tillage and the Environment’ was published (10,000 copies). APDC now works closely with Embrapa, the Ministry of Agriculture and Supply and with the Water Resources Secretariat of the Brazilian Ministry for the Environment and Amazônia. The latter has recognised the importance of ZT as a watershed management tool and is funding APDC efforts to create and support FSC’s. 

In 1998/9 the planted area in Zero Tillage for the Cerrado region is expected to top 4 million hectares, official research is now giving high priority to Zero Tillage and extension services have indicated a wish to train their personnel. According to Santana et al., 1994 (data from one rainy season planting with field equipment) erosion levels under zero tillage drop to between 5 and 12% of those with CT, while rainfall infiltration is increased by 30 - 60%. Zero Tillage also encourages maximum soil biological activity and is highly positive for the environment, representing the best alternative for watershed management in cropped areas. Derpsch (1997) cited American work to show the notable effect of carbon sequestration in ZT, reducing the liberation of CO₂ to the atmosphere. ZT is showing high potential in both the wet/dry and the humid tropics and should be recommended as the sole system for cultivation in the Amazon region (Landers, 1997). 

The Phases of Development 

I. The Pioneering Phase - 1981 to 1986 

• isolated pioneer farmers testing new ideas;
• insignificant government research;
• only adapted planters with wavy disc coulters;
• very high cost of contact and dessicant herbicide and poorly defined application rates;
• ineligibility for credit
• no teaching of ZT;
• no adoption of ZT by extension agencies (no official recommendation).

II. The Consolidation Phase - 1986 to 1992 

• private sector R&D programmes
• resolution of basic agronomic problems of second-cropping. dessicant application rates, crop rotations, improved planters;
• herbicide costs falling and machinery and credit costs rising;
• introduction of eccentric double disc coulters (openers);
• commercial fertilizer and herbicide firms began to stimulate development of Zero Tillage technology in various ways;
• beginning of dissemination of early positive R&D results through field days, short courses and individual contacts;
• pilot Bank of Brazil credit program for second crop ZT.

III. The Mass Action Phase - 1992 to 1997 

• a wide range of agronomic solutions available for sustainable systems in the humid and wet/dry tropics;
• direct costs of ZT equal or lower than conventional tillage;
• very rapid expansion from <200,000 hectares in 1992 to na estimated 3 million ha. in 1997/8 crop season;
• foundation of the Associação de Plantio Direto no Cerrado (APDC), a farmer/technician NGO;
• much-improved planters, with introduction of the guillotine knife coulter;
• widespread adoption of ZT by private technical assistance and co-operatives;
• direct support from private sector for farmers associations, field days, seminars, technical publications;
• little formal inclusion in university and college curricula.

IV. The Dominant Phase - 1998 to 2100 

• research priority for ZT, solving ZT’s second generation plant pathology, soil fertility/physical, machinery and variety selection needs;
• incentives for ZT adoption;
• adoption by extension services;
• inclusion in university and college curricula.

Erosion Control with Zero Tillage 

Table 1: Model Estimates of Tropical Soil Losses by Erosion Under Conventional Tillage 

Source : Chaves, H.M.L. 

Landers (1996) cited figures from a personal communication of H.M.L.Chaves for mathematical models of erosion levels under conventional tillage for part of the São Francisco river basin in Bahia state, shown below, which gave a 70% efficiency when correlated with measured silt levels and a weighted average soil loss of over 9 ton/ha for the watershed in question. Losses from purely arable areas would be significantly higher. 

Control of erosion with zero tillage systems has been dramatic. Resulting reductions in soil losses and surface flows can be evaluated in the figures of Santana et al ( 1994), Table 2. Although partial figures only were available, the impact of ZT is evident, with average soil losses between 5 and 12 % of conventional systems. Dedecek et al. (1986), comparing CT with simulated zero tillage, using a hoe for planting and fertilisation, showed lower differences in erosion losses and runoff but still showing considerable advantage for ZT, including reduced losses of soil nutrients. Besides the physical effects of erosion, planting delays to prepare soil during heavy rain periods and re-planting costs for eroded areas are severe with conventional tillage using disc implements. 

Table 2: Losses of Soil and Water in Cerrado Latosols, Sete Lagoas, MG, Brazil Under Different Methods of Soil Preparation 

Fonte: Santana et alii (1994). 

Merten (1993) in the sub-tropics has shown erosion losses under animal traction ZT to be about 10% of the traditional ploughing and 20% of minimum tillage losses with scarification. It is expected that a similar advantage will be shown for the tropics. 

Impact on the Environment 

By virtually eliminating erosion effects, ZT permits sustainable farming in tropical latitudes; it also has a number of far-reaching indirect impacts on society with little government investment. The box below summarizes these impacts: 

Box 1. Benefits Generated for Society by Adoption of Zero Tillage. 

• Reduction of silting in reservoirs, lakes and watercourses proportional to 70-90% less erosion (Chaves H.M.L in Saturnino and Landers, 1997) - a very conservative estimate of the annual value for the Cerrado region was given as 33 million US dollars per year (Landers, 1996);
• Consequent reduction in the pollution and eutrophication of surface waters by agricultural chemicals carried in erosion runoff (Sorrenson and Montoya, 1984);
• Substantial reduction in treatment costs of municipal water drawn from surface sources (Braganolo and Parchen, 1991);
• Considerable reductions in maintenance costs of rural roads;
• Reduced wear on hydro-electric turbines from the passage of cleaner water;
• Flooding risks are reduced by 30-60% greater rainfall infiltration (Chaves,H.M.L in Saturnino and Landers, 1997) and delay to overland flows by surface residues, increasing times of concentration;
• By the same token, aquifer recharge is enhanced, improving groundwater reserves and dry season flows in springs and streams;
• Reductions in diesel fuel of 50 to 70%, or more, (Gentil et al., 1993) and proportional reductions in greenhouse gas emissions;
• Zero Tillage per se has a major impact in reducing carbon dioxide emissions when compared to Conventional Tillage, by immobilizing carbon in incremental soil organic matter and surface residues (Derpsch, R. in Saturnino and Landers,1997)
• It is axiomatic that, by promoting high-yielding sustainable agriculture and increasing pasture carrying-capacity, through rotation with annual crops, Zero Tillage takes pressure off the demand for agricultural frontier expansion by deforestation;
• Provision of winter feed (crop and weed seeds not incorporated) and shelter, lower soil temperatures and reduced water pollution levels increases populations of terrestrial, soil and aquatic fauna;
• a high-yielding, prosperous and sustainable agriculture ensures lower food costs and improved food security for the population as a whole.

Source : adapted from Landers J.N., 1994. 

Luchiari Jr., A. et al.(1997) showed a reversal in the upward trend in use of water treatment chemicals after the introduction of Zero Tillage to an expanding and intensively irrigated area in northern São Paulo state and Braganolo and Parchen (1991) showed reductions in cost of water treatment chemicals and in the maintenance of earth roads associated with Zero Tillage, while Sorrenson and Montoya (1984) showed high levels of solute content in the water of Itaipú reservoir coinciding with the periods of conventional land preparation in its drainage basin. 

Financial Analysis 

Improved financial performance is of paramount importance as a motive for adoption of ZT. However, although there may be some gains in direct costs, in the order of 5-10% for advanced farmers, the decision to adopt also takes into consideration a number of secondary factors which have financial and management implications. The analysis below of a large mechanized dryland operation planting 2 years soybean to one year of maize indicate a substantial advantage of ZT as compared to CT. In the medium term, the effect of reductions in machinery replacement is an important factor in the higher return of ZT. 

An APDC farm survey showed an average reduction of 44% in tractor investment, expressed as hp/ha, with dryland indices for ZT ranging from 0.26 to 0.40 hp/ha and 0.32 to 0.54 hp/ha for irrigated conditions (Landers, 1994). A single farm survey on a 2110 ha farm showed showed 10,662 machine hours for the year 1992, before ZT and only 4,761 hours under total ZT in 1996. This example also illustrates the very rapid adoption, reflecting confidence in the technology and the rapid growth of ZT area in the Cerrado region; Gentil et al (1993) calculated economies of between 64 and 74% respectively in diesel fuel consumption for theoretical studies of 500 and 2000 ha. operations. 

Landers (1994) and Gentil et al (1993) have shown reductions in prodction costs for ZT of up to 13%, depending on the crop and scale of the operation. 

Table 3: Financial Analysis Comparing Internal Rates of Return of Zero Tillage with Conventional Tillage at a 10% Discount Rate. 

Source: Landers,. et alli (1994) 

Not taken into account were the intangible effects of more time to take management decisions and to upgrade technical knowledge and greater satisfaction in feeling that the system appears to be really sustainable and in harmony with nature. The sea-change in farmers’ thinking patterns is illustrated in Figure 2. 

Figure 2: A Management Route to lower use of agricultural chemicals (Source: Adapted from Landers (1996) 

Effects on Soil Properties 

As indicated earlier, CT, especially in tropical latitudes, provokes rapid oxidation and hence declining levels of soil organic matter (Silva et al 1994 , Resck et al 1991). This is reversed by zero tillage; the resulting accumulation of soil organic matter and carbon being more notable in temperate and sub-tropical areas than in tropical regions. Freitas (1995) observed that soil infiltration rates and fertlizer efficiency were enhanced by ZT and that scarifying soil to remove plough pans was only a temporary solution. 

Changes in Fertilizer and Soil Correction Practices 

Losses from surface applied urea on tropical oxisols under ZT conditions and with high residue levels can reach 78%. Incorporation of urea to a depth of 5 to 8 cms is a sine qua non for N efficiency and reduces losses to the order of 5% while irrigation immediately after application and liquid mixture with sulphate of ammonia or potassium chloride also mitigate N losses (Lara Cabezas 1997). This practice minimizes losses by volatilization as a result of urease liberated by oat straw and N sequestration by microorganisms digesting surface mulch. The Friends of the Soil Club from Rio Verde in Goiás state demonstrated that application of total N incorporated at planting (separate from the seed) or 30 kg/ha of N at planting and 90 kg/ha topdressed at 15 days after emergence gave better results than the traditional application of the same amount of topdressed N at 35 days after emergence (Ferreira, 1996). 

Sá (1993) showed that surface application of lime after 270 days was superior to incorporation to 20 cms depth, under ZT on distrophic red-yellow and dark red latosols in Paraná state (sub-tropical). While farmers have proved the validity of this practice for tropical latosols, Brazil’s tropical research still lags in recommending the technique ; Landers(1993) made a provisional recommendation of 50% of Sá’s recommended levels in order not to provoke micronutrient deficiencies, through immobilization as a result of an elevated p.H. in the high organic matter surface zone. Farmer experience corroborated this position when 3 ton/ha of dolomitic lime, surface applied, provoked Manganese deficiency, which required aerial spraying to correct the deficiency (Tonon, 1995 personal communication). 

De Sousa, (1996) recommended minimal levels of soil P₂O₅ as a pre-condition for broadcast application of Phosphate fertilizer under ZT conditions as follows: 

Table 4: Minimum P₂O₅ Levels for Adoption of Broadcast Application of Phosphate in ZT. 

Increasingly, the concept of high base saturation, as a pre-condition for top tropical soil yields is being questioned, especially in the ZT situation (Altmann,N. personal communication, 1998). In addition, it has also become clear that free Aluminium in the soil solution, as a negative factor for plant growth, is a result of its liberation by oxidation of soil organic matter and that, in as far as ZT reduces or avoids this, high yields may be attained with base saturation levels of only 30-50% (Sá, J.C. de M. personal communication, 1998). And, further, soil analyses which measure total and not free Aluminium do not give the true picture. Corroborating this is a farm-level soybean yield of 2.640 kg/ha with only 150 kg/ha of slaked lime in the row on an Amazon latosol at 12ºS(Séguy et al., 1997) on land cleared with a minimum burning technique which preserves up to 70% of the original organic matter in the system (Zuanazzi, A.. personal communication 1997). 

Farm-level tests show that there is considerable margin to reduce soybean fertilization rates, but the degree to which this can be imputed to ZT is unclear, however, Sá (!993) noted significant reductions in phosphate fertilization on sub-tropical latosols. 

The Importance of Crop Rotations for Sustainability 

The generation of adequate levels of crop residues to cover the soil until a full crop canopy has formed is essential to ensure sustainability. The inclusion of maize (corn), rice or wheat and/or specialised cover crops for this purpose has been essential in rotations designed to break the cycles of weeds and pests and to compensate for higher value crops with inadequate quantity and quality of residues, such as edible beans and soybeans, with high C:N ratios in their straw. 

Double cropping opportunities depend on the regularity of rainfall in the months of February to May and the date of harvest. Farther West the rainfall is more dependable in these months and even maize is double stopped in SW Goiás and Mato Grosso up to about mid-February, after early 105-115 day soybean varieties. Grain sorghum is the next most important second crop, with millet following as a last option. The latter two crops have been successfully oversown into maturing soybeans at, or just before, leaf fall (Landers, 1994). 

This technique most commonly uses aerial seeding on large areas or a pendular fertilizer spreader on smaller ones - the relatively high risk is offset by very low cost of US$6-8/ha plus own seed cost at about US$0,1/kg for 25-30 kg/ha. 

Gradually, as transport conditions and market demand from intensive livestock enterprises have improved, more maize has been grown, in rotation with soybean - at the present time the proportion does not attain one third of total farm acreage. This not only improves residue levels but assists in weed and pest control, plus improving yields of both crops Ruschel (1992). High transport costs, low local demand and higher temperatures limit maize area in northern of Mato Grosso, where high milling quality upland rice is the preferred gramineous crop in the rotation, giving farm yields of 3 to 5 ton/ha.(Séguy, Bouzinac et al 1993). 

Vasconcelos and Landers(1993), Landers(1994) and Séguy, Bouzinac et al (1993) report on experiences of planting grain crops into permanent cover crops, of which, maize into Siratro was the most successful, giving a greater margin than conventional maize, chiefly through the elimination of the selective maize herbicide (Vasconcelos and Landers 1993). De Sousa (1996) reports that planting maize directly into the Mineirão cultivar of Stylosanthes sp. gave yields of about 6 ton/ha and Ayarza et al.(1997) demonstrated the weed-smothering capacity of an Arachis pintoi swards as being equal to control with herbicide but having the same effect on ZT maize without herbicide; however this species is slow to establish and seed production is expensive. However, with herbicide the ZT maize produced some 7,5 ton/ha under ZT, superior to CT with herbicide, but less than the treatment with a subsoiler, which produced 8 ton/ha. less than the treatment However, these systems require a very high management level and have not been adequately tested for widespread use. 

Integration of crops and livestock 

Planting soyabean directly into Brachiaria spp. has become commonplace in the Cerrado region. Dessicant dosage rates vary from about 2 l/ha of glyphosate (commercial product) on B. ruziziensis to over 4 on B. brizantha, with B. decunbens at an intermediate level. After pre-plant dessication, the planter l/ha breaks through the superficial compaction (8-10cms) of the pasture and plants soyabean into the dead/dying sward. Broch et al (1997) at ^o S report yields of 3500 kg/ha of soya in the first year, maintaining productivity of over 3000 kg/ha in the third year. Beef production dropped more sharply, from 375 kg/ha/annum in year 1 to about 120 kg/ha in year 3. 

Soil Cover and Cover Crops 

By the time ZT was introduced to the Cerrado region, there was a sufficient reservoir of basic knowledge, transferred from South Brazil, to serve as a basis for the pioneers to start using the new technique with relative success, attracted mainly by its capacity to control erosion. The systemic dessicant herbicide, glyphosate was already available as were kits of trash discs for adapting planters to ZT. The principal agronomic limitation to the system was maintenance of soil cover. Approaches used were : 

(i) utlization of volunteer weed cover; 

(ii) use of double cropping, principally with sorghum and millet; 

(iii) increasing maize into the soybean monocropping system; 

(iv) introduction of a permanent green undercrop. 

The first solution was gradually substituted for millet (Pennisetum typhoides), including the two new private sector varieties BN1 and BN2 specially selected for Zero Tillage Bonamigo ( 1993) which has the advantages of suppressing troublesome perennizing weeds (Sida ,Tradescantia and other spp), producing more even resistance, which facilitates planter adjustment. About one million hectares of millet are planted for soil cover and eventual use as a fodder or catch crop for grain or seed. Millet is by far the most widespread cover crop planted today in Brazil’s tropical regions. It suffers the disadvantage of acting as trampoline for army worm (Spodoptera fugiperda) populations, which then attack maize or other crops; so far serious diseases have not attacked millet. 

Crops produce differing amounts and qualities of residues. A wide C:N ratio and higher lignification level reduce the velocity of dry matter oxidation. This is clesrly shown in Table 5. below. 

Table 5: Evolution of the Loss of Dry Matter of Crop Residues and of Percentage Soil Cover. Fazenda Progresso, Lucas do Rio Verde, MT, Brazil 1985-1989 

Source: L. Seguy, S. Bouzinac et alii (1992). 

The Effects of Mulch on Weed Control and Soil Moisture Conservation 

A number of farmers are reporting reduced levels of post-emergent herbicides on soybeans planted into heavy millet residues, in a few cases, to zero. This is in part due to suppression during the cycle of the millet, but also to effects of the heavy mulch. These may be due to smothering or allelopathy, or both. Allelopathic suppression from sunflower residues and of rice straw on new rice seedlings have also been noted. 

The graph below (Stone and Moreira, 1998) indicates water economy in irrigated beans of up to 40% at the same yield level for the "Safira" variety, which is an upright typeThis economy is lower for prostrate varieties with higher soil cover over time. 

Figure 3: Yield of the Bean Cultivar Safira as a function of the Depth of Water Applied, with Three Methods of Soil Cultivation G = offset disc, PD + Cob. Mort = Zero Tillage with (Added) Mulch Cover and PD = Zero Tillage 

A similar water economy effect pertains under dryland conditions (Landers, 1994), when yield differences can be dramatic after 20-25 day droughts in the growing (rainy) season. 

Control of Red Rice in Flood-Irrigated Rice 

Séguy and Bouzinac (1993) at 4^o S, in Piaui state, obtained almost total control of red rice using pre-plant dessication of previously germinated red rice seedlings with glyphosate and drilling directly into the dying weeds. The number of red rice panicles per square meter was reduced from a range of between 52 to 79 plants per square meter under conventional tillage rice monoculture to less than one with ZT and a rice/bean rotation. The „Minimum Tillage plus Direct Drilling" technique has been plant per square meter in normal use in sub-tropical Brazil for about ten years on extensive areas. 

Pests and Diseases 

In general, in the early years of ZT some problems are encountered, partly due to farmers not being alerted to early warning symptoms. In mature Zero Tillage, biological controls are maximised. 

Perhaps the greatest problem is found with slugs under irrigated conditions. Baits with metaldehyde. Or carbaril and maize meal plus beer or radish juice as an attractant have been effective but labour intensive, since rain resistant pellets are not yet on the Brazilian market. The species involved have not been identified but are effectively killed by these baits. 

Large white scarab beetle larvae of the Phytalus and Phyllofaga geni have caused sporadic and patchy damage to crop roots in early years of ZT, but this has not continued. Another genus, Buthinus creates vertical tunnels to 40 cms or more and hole numbers have exceeded 20/sq.meter, greatly improving infiltration, to the point where farmers are lowering or taking out contour banks after five years or more of ZT. This genus feeds on crop residues and only attacks plants in their absence (personal communication J.C.Rauber). Ants and termites of various species tend to increase with ZT and control measures must be more vigilant. Termite mounds of uip to 1 meter high can appear in one season and pose a problem for combine harvesters.Armadillos are attracted by the Scarab larvae and tend to concentrate on isolated ZT fields. 

Stem canker of soybean ( Diaporthe phaseolorum) no longer poses a threat because resistant varieties have been made available; scientists differed on whether ZT affected the disease or not (Yorinori, 1996, Nasser.L.C. personal communication). Several late season disseaes of maize and soybean are becoming more prevalent; whether this is due directly to ZT or indirectly, as in second crop maize propagating inoculum, is not yet clear. Fusarium root rot in soybean and beans is increasing, but White Mould ( Slerotinia sclerotium) in Phaseolus beans appears to be less prevalent under ZT conditions due to natural control mechanisms (Nasser 1998). 

Perennial weeds tend to increase and require careful monitoring and rotation of crops and herbicides. Post harvest dessication is showing promising results for their control. 

Planting and Spraying Operations 

Planting should be on a recently sprayed cover crop or on one which has completely dried. Wilted straw causes „hairpinning", when the straw doubles up under the disc and prevents good seed-to-soil contact. There is a similar limitation on very wet soils, when the lack of resistance against which to cut the straw causes the same effect. Combines should always be equipped with straw spreaders for easy planting. Shredders use much power and the fine pieces decompose faster. A Brazilian vertical rotary chopper is sometimes used on excessive straw burdens to hasten decomposition before planting.Weed cover only tends to give very uneven residues and makes planter adjustment difficult. 

Brazilian ZT planters have the following mechanisms : 

1. Double disc coulters with one axis leading slightly on the other for seed and fertilizer;
2. Double disc coulters with one axis leading slightly on the other and paired discs of different diameter for seed and fertilizer;
3. In heavier trash situations, a single striated or smooth cutting disc may be mounted in front of these coulters ; some planters have a slight angle on these discs, to help cutting, while wavy discs have become obsolete because they disturb the residue excessively;
4. A fertilizer knife coulter mounted behind a single smooth cutting disc, with a forked „guillotine" disc cleaner mounted between them - this must be constantly adjusted for wear as the clearance is very critical;
5. An alternative is to mount the knife some 30 cms or more behind the trash disc;
6. A farmer adaptation is to use a piece of pickup spring to make a fine in-row tine between paired eccentric double discs for fertilizer and seed which leaves the straw more or less intact on the surface.

In many situations and always with maize, there is a positive response to an in-row knife or tine. Reducing water applied in spraying gains significant sprayer time and helps absorption of systemic dessicants. In the tropics, 100 litres/ha is now common as opposed to 30litres/ha some years ago. 

Marking for pre-plant spraying requires different techniques from CT and combines should be equipped with straw spreaders. 

The process of technology transfer 

The principal mechanisms involved in the adoption of ZT in the „Cerrado" wet/dry savannah region of Brazil were : 

• farmer-to-farmer contact
• commercial and NGO sponsored events
• technical assistance/promotion activities of herbicide firms
• private and cooperative technical assistance
• NGO/private sector publications
• media reports

The special case of municipal-level „Friends of the Soil" clubs is an example of farmer self-help in the absence of adequate government technical or credit support. These are non-profit, non-commercial and non-political entities, affiliated to APDC, which gives technical and operational support. They have regular monthly meetings on topics of interest and are highly effective both in spreading the technology and in raising the level of farmer achievement. Their activities, which readily gain supplier support, are summarized in the box below : 

Box 2. A New Model for Technology Generation and Transfer 

Source : Landers (1998) 

Lessons learnt 

(i) Local „Friends of the Soil Clubs" take on a missionary zeal and are very effective in farmer-to-farmer technology transfer (ii) farmer innovations and private sector R & D were much more important than government research (over which farmers had no control) forging the new technology (iii)after the initial euphoria of conversion to ZT there (iv) partnerships with input and machinery suppliers are easily developed for technology transfer to farmers which has immediate impact on their commercial demands (v) zero tillage is working in all wet/dry and humid tropical situations (vi) the change to „think zero tillage" is a 180º turn and farmers take time to adjust, so conversion to ZT should be over a period of 4-5 years (vii) the model of a regional farmer/technician NGO articulating with private sector and government to give support to local farmers’clubs has worked efficiently (viii) lack of some farmer control over research priorities impedes sharp focus on immediate production problems. 

Areas for Further Research 

ZT for Amazonia. There is an urgent necessity for research to develop ZT and zero burning practices for conversion of degraded pasture in the Amazon region into productive crop land, rotated with high carrying capacity pastures as a means to discourage deforestation. A start has already been made with farmers in northern Mato Grosso (Séguy et al 1998) and Kanashiro, M. at Embrapa-Humid Tropics Centre, Belém, PA. 

Planters and drills. Throughout the Cerrado region, according to many farmer interviews, planter/drill performance suffers from lack of adaptability to varying soil, residue and weather conditions and short life spans of working parts (Landers, 1998) 

Insect Pests. The dynamics of soil insect populations in the initial phases and under long term ZT require elucidation, especially the Scarabidae family beetles, slugs, ants, termites, grasshoppers/ locusts and crop pests such as Sternechus subsignatus on soya. 

Diseases. The presence of surface humus favours increases in the inoculum of certain diseases but assists in controlling others, such as White Mould (Sclerotinia) in Phaseolus beans. The dynamics of these diseases and of antagonic fungi need further clarification to avoid farmers reverting to ploughing as a (sometimes spurious) control measure (da Costa, 1998). 

Cover Crops Widescale screening of world collections of possibly interesting species is required to substitute isolated trials with limited materials and non-comparable experimental methods. 

Soil compaction The effects of surface residues on wheel compaction and compaction problems in Quartz Sands in the NW Bahia region require basic research. Research is required to solve rooting problems in some ZT situations and the value of shallow scarification as a solution, including the design of tines for shallow operating depths and drier soils (to reduce clod formation). 

Variety Selection National trials under ZT conditions. 

Liming and fertilization Re-calibration of liming and nutrient recommendations under ZT conditions, especially Phosphorus. Planting techniques into undisturbed natural soil and pastures and the interaction of organic matter with aluminium toxicity vis-à-vis liming requirements. 

References 

^{Executive Secretary, Associação de Plantio Direto no Cerrado, Brasília, DF., Brazil.} 

MUDB, Conjunto 9, Lote 5. CEP71680-090. E.mail: john.landers@apis.com.br