Management of irrigated wheat
K.D. Sayre

In the hills of Nepal, most wheat is managed entirely by direct human labour, from land preparation and planting to harvesting and threshing. On occasion, land preparation may be assisted by the use of bullocks, but wheat production essentially represents the intimate involvement of the farmers and their families in the wheat field performing the necessary management tasks. At the other extreme are the highly mechanized areas of wheat production in countries such as Argentina and Brazil where land preparation, planting and harvesting is primarily done by machine. Many other areas represent intermediate situations where draft animal power, mechanization and human labour are combined to produce the wheat crop. In all cases, however, farmers strive to maximize their available production resources while minimizing risk. This is the essence of crop management and agronomy.

WHEAT ENVIRONMENTS

Wheat is the most widely grown crop in the world and is produced under a wide range of environments (Hanson et al., 1982). The delineation of these production environments is largely conditioned by temperature and moisture availability (via rainfall and/or irrigation). The International Wheat and Maize Improvement Center (CIMMYT) has defined 12 mega-environments (for description see chapter "CIMMYT international wheat breeding") that are derived from combinations of moisture conditions (irrigated, high-rainfall or low-rainfall moisture contrasts), prevailing temperatures (temperate, hot or cold extremes) and prevalent biotic and abiotic stresses (Rajaram and van Ginkel, 1996). Table 24.1 summarizes the main agro-ecological zones where wheat is grown in developing countries around the world.

By far the most important agro-ecological zone for wheat production in the developing world is the temperate, irrigated condition characterized mainly by large areas in northwest India, Pakistan, Nepal and northern Bangladesh, south-central China, Afghanistan, parts of West Asia (mainly in Iran, Turkey, Syria, Iraq, and Saudi Arabia) and the central Asian republics. Asia now produces over 90 percent of irrigated wheat in the developing world. There are smaller, yet important temperate, irrigated wheat production areas located in North Africa (especially Egypt), Zimbabwe, South Africa, northwest Mexico and Chile.

Areas where wheat is grown under high-rainfall, temperate conditions include the Southern Cone countries and parts of the Andean region, Ethiopia and some areas in North Africa and in the Himalayan region of Asia. Production areas with temperate, low rainfall include North Africa, West Asia, parts of Pakistan and Afghanistan, northern India and north-central China, as well as some areas in Latin America including the central Mexican highlands.

Examples where wheat is grown under high temperatures (with or without irrigation) include the Sudan, Nigeria and east African countries such as Uganda and Tanzania, parts of Bangladesh and India and large areas of Brazil, Paraguay and southeast Bolivia.

The developing country where the most wheat is grown under high latitude is China, and likewise China produces a great deal of the developing world's facultative and winter wheat along with Turkey, Iran and many of the central Asian republics.

TABLE 24.1
Principal agro-ecological zones for wheat and their relative contribution to developing country wheat production

Source: CIMMYT, 1989.

A table similar to Table 24.1 but showing the distribution of these same agro-ecological zones for developed countries would be quite different. Less that 5 percent of the developed world's wheat is grown under irrigated conditions (CIMMYT, 1989). Large wheat production areas in the United States, Canada and Australia are also produced under low-rainfall conditions, whereas most wheat production in Western Europe is produced under favourable rainfall conditions. As a further contrast, winter wheat is by far the most common type in developed countries, although most wheat in Australia and sizeable areas in the north-central United States and in Canada are planted to spring wheats.

This contrast, in which wheat is being produced essentially as an intensively managed crop (especially where irrigation is used) in most developing countries versus its more extensive management under rainfed conditions in many developed countries, has obvious agronomic implications. Emphasis in this chapter, however, will be given to the discussion of wheat management practices in the developing world, stressing the temperate, irrigated spring wheat production systems.

IRRIGATED WHEAT CROPPING SYSTEMS

There are a multitude of cropping systems that occur where spring wheat is grown under temperate conditions with irrigation, and by far the largest such wheat production areas are found in South and East Asia. For example, India, Pakistan, Bangladesh and Nepal have a total wheat area of over 34 million ha (CIMMYT, 1996). Over 12 million ha in these countries are devoted to the rice-wheat system with about 9 million ha irrigated (Hobbs and Morris, 1996). Of the remaining 22 million ha of wheat in South Asia that is not grown in rotation with rice, a large but poorly documented area is also grown with irrigation in rotation with various upland crops including cotton, soybean, sugar cane, maize and sorghum, as well as other minor crops.

Similarly, China has over 29 million ha of wheat with close to 50 percent of the area irrigated (CIMMYT, 1989). The rice-wheat rotation plays an important role comprising 35 percent of the area planted to wheat and with nearly all under irrigation. In other areas in China with irrigation, wheat can be found growing in rotation (in many cases as relay crops) with soybean, maize, sugar beet, sun-flower and many other crops.

In many countries, such as Mexico, Chile, Egypt and Zimbabwe where wheat is also grown under temperate conditions with irrigation, main rotation crops are maize and soybean with smaller areas in rotation involving cotton.

CULTURAL PRACTICES

Tillage and planting

Irrigated wheat production systems in developing (and developed) countries depend almost entirely on conventional, intense tillage practices, especially where irrigation is by gravity delivery to the field via flood irrigation or by furrows. It is commonly assumed that modern, sustainable crop management systems are characterized by reduced or zero-tillage prior to planting combined with retention of the previous crop residues; however there are very few applications anywhere of such practices in gravity-irrigated wheat production systems, which as seen in Table 24.1, constitute the most important wheat production situation in the developing world. In nearly all cases, multiple tillage operations are used to prepare the seedbed and to allow a levelling operation with the construction of borders for flood irrigation or the formation of corrugations for furrow irrigation.

In most of Asia, the previous crop residues are normally removed for livestock fodder and/or for use as cooking fuel. In some cases, they are left and incorporated during the tillage operations or may be partially returned to the field and incorporated via compost or farmyard manure. However, burning of crop residues is becoming more common as more farmers begin to use combine harvesters, especially when crop turnaround time is short and the time required to incorporate the normally high levels of residues can lead to major next-crop planting delays. However, declining soil organic matter levels and degradation of associated soil physical parameters in many soils, particularly in many areas characterized by continuous irrigated rice-wheat production (Hobbs and Morris, 1996), argue for the development of new crop management strategies and perhaps new public policies that discourage the burning of crop residues both in the field and for cooking fuel.

Crop residue yields can be quite high in irrigated wheat systems. Rice straw yields can range from 5 to 9 tonnes/ha in irrigated rice-wheat rotations in South and East Asia, while maize residue yields can be in the same range in irrigated wheat-maize rotations in northwest Mexico. These are double to triple the residue levels normally associated with most zero-tillage systems with residue retention in low-rainfall, rainfed wheat production systems and are a potential major obstacle to implementation of zero-tillage planting systems if all or even partial amounts of the crop residues are to be retained on the surface in gravity-irrigated wheat production systems.

In Zimbabwe, there are farmers who grow high-yielding wheat (up to 9 tonnes/ha) in rotation with high-yielding maize (up to 12 tonnes/ha) or soybean (4 to 6 tonnes/ha) under reduced and/or zero-tillage but with sprinkler irrigation systems who still encounter difficulties when planting into such high levels of crop residues if they attempt to retain all residues in the field. These farmers do not have to confront the potential problems associated with surface irrigation in the presence of residues. Seed placement and stand establishment in such high levels of crop residues offer challenges for the development or adaptation of existing reduced- or zero-tillage planters.

Over the past 15 years, researchers in northwest Mexico together with farmers have developed and implemented what is fundamentally a new system for planting wheat. As a result, most wheat is no longer planted as a solid stand on the flat with flood irrigation, but rather on raised beds (usually 15 to 25 cm high) that are normally 70 to 90 cm wide between furrows (similar to row widths for other widely grown crops in the area, such as maize, soybean, cotton, sorghum and safflower), with two to three individual rows on top of the bed spaced 15 to 30 cm apart.

The furrows between the beds are used for irrigation water delivery. The planting system is different from the furrow-irrigated systems occasionally found in some irrigated wheat areas where the wheat seed is broadcast or drilled on the flat followed by the formation of the irrigation furrows at approximately 90 cm intervals. The system used by the farmers in northwest Mexico is unique in that the wheat is only planted in defined rows on top of the bed. This planting system allows farmers greater flexibility in managing their wheat crop as compared to conventional solid seeding on the flat. Below are listed several reasons for the wide adoption (nearly 90 percent of the farmers in the state of Sonora, Mexico) of the bed-planting system for wheat that farmers in northwest Mexico have expressed through on-farm surveys:

• improves irrigation water management;

• facilitates pre-seeding irrigation, thus providing opportunity for initial weed control prior to planting;

• allows better stand establishment;

• makes it possible to perform inter-bed mechanical weed control during early crop cycle;

• uses lower seed rate;

• reduces crop lodging;

• same beds can be reshaped for planting the successive soybean or maize crop (usually with crop residue burning);

• reduces herbicide dependence;

• facilitates hand weeding.

These periodic surveys have indicated that herbicide use in wheat has declined from a high of nearly 60 percent farmer use in 1981 when only 6 percent of farmers planted wheat in beds to just 25 percent in 1996 when nearly 90 percent of farmers were planting wheat in beds (CIMMYT Economics Program, personal communication, 1996).

Almost no farmer indicated that adoption of the new planting system was deliberately done to obtain higher yields specifically, but involved decisions to increase production efficiency and reduce production costs. However, a survey conducted in 1994 in the Yaqui Valley in the state of Sonora, Mexico, showed average wheat yield for farmers growing wheat on beds at 5 615 kg/ha whereas, those farmers still planting in solid stands on the flat with flood irrigation averaged 4 923 kg/ha (Aquino, 1998). With experience gained over time, farmers in the Yaqui Valley have learned how to obtain superior wheat yields while benefiting from the economic advantages from bed-planting.

Research has demonstrated that not all wheat cultivars perform well with the bed-planting system and part of the current higher yields found in farmers' fields with bed-planting in the Yaqui Valley can be attributed to a concerted breeding effort to identify new cultivars better adapted to bed-planting. Table 24.2 presents the results from an experiment conducted at the CIANO experiment station (located in the Yaqui Valley) under high yield management to compare several bread wheat cultivars under conventional planting (seeded on the flat with flood irrigation) versus two different bed-planting systems.

TABLE 24.2
Comparison of grain yields for conventional planting on the flat versus bed-planting at high and low seed rates

^a LSD (0.05) for genotype mean is 395 kg/ha; interaction LSD (0.05) is 684 kg/ha.
^b Grain yield kg/ha at 12 percent H₂O.
^c Means followed by the same letter do not differ significantly at LSD (0.05).

Clearly cultivars such as Oasis 86 and Super Kauz 88 appear not well-suited for bed-planting, whereas Baviacora 92, CIANO 79 and WEAVER 'S' show high and stable yields over the three planting systems (note the minimal effect of seed rate and number of rows of wheat per bed on yield). Obviously, breeding for appropriate cultivars must proceed together with the development and implementation of new crop management strategies, such as planting wheat on beds.

Planting wheat on beds also opens up new opportunities: (i) to improve the efficiency of fertilizer use (especially nitrogen) because of the easy access to enter the field by use of the furrow space between the beds that enables the placement of nutrients when and where the wheat plant can make more ready use of them; and (ii) to reduce tillage, manage crop residues and control field machinery traffic by using the same bed for successive crops (permanent beds) while confining machinery wheel traffic and irrigation water movement to the furrows between the beds.

Research was initiated in 1992 by CIM-MYT wheat agronomists in the Yaqui Valley to develop a feasible irrigated, permanent bed technology to plant wheat in the winter cycle and either maize or soybean in the summer in two-crop annual rotations. Considerable effort has been applied to modify existing cultivators to allow reformation of the beds after the harvest of each crop in the presence of retained crop residues without disturbing the top of the bed where succeeding crop is seeded. Similarly, adaptations of existing planters have been made to allow planting two or three rows of wheat on top of the bed without tillage and into residues. Many of the machinery modifications that have been developed borrow from the ridge-tillage conservation tillage system used primarily for maize and soybean in parts of the north-central Corn Belt in the United States. However, more attention is being given to controlling distribution of the chopped crop residues either in the furrow or on top of the bed or both depending on the crop being planted. Traditional ridge-tillage usually whips off all crop residues into the furrow between the ridges (or beds).

Table 24.3 presents the wheat yields averaged over four years (from 1993 to 1996) from an experiment in which wheat has been continuously grown in an annual rotation with summer maize since 1992. The experiment involves different tillage and crop residue management practices as well as different rates and timings of nitrogen (N) applications (applied as urea), which are explained in Table 24.3.

Tillage/residue management means differed significantly, with all permanent bed treatments significantly outyielding the conventional-tillage treatment. All tillage/residue management treatments showed significant responses to N rates up to 225 kg N/ha and some responded up to 300 kg N/ha. The tillage/residue management by nitrogen management interaction was significant, and the first node N application was very effective for treatments where residues were retained.

It appears that the innovative approach to planting wheat on beds offers new opportunities for farmers to manage irrigated wheat crops more efficiently and may lead to a technology that allows major reductions in tillage combined with crop residue management in surface-irrigated production systems.

TABLE 24.3
Wheat yields averaged over four years for tillage/residue management and nitrogen management treatments on beds, Yaqui Valley, Sonora, Mexico

^a Conv. = conventional tillage with new beds formed for each crop; Incor. = residue incorporated; Reduce = permanent beds, reshaped between each crop; Burn = all residue burned after harvest; Partial = removal of wheat straw that was cut by combine (15 to 28 cm stubble left); Remove = residue raked, baled and removed; Retain = all residue chopped and left in the field.

^b Means followed by the same letter do not differ at LSD (0.05); LSD (0.05) for interaction is 408 kg/ha.

Stand establishment

Obviously, establishing an adequate wheat stand is critical for achieving high yields. Wheat, however, because of its growth habit and morphology is able to compensate to some degree (especially under optimum, irrigated conditions) for less than optimum stands through the cardinal yield components of tillers per plant, grains per spike and grain weight. Cultivars differ in their ability to compensate for poor stand and also differ in the relative importance of which the yield component acts to compensate the poor stand. Usually, increasing the number of tillers per plant plays the most important role, but some cultivars show large differences in grains per spike under different conditions. Normally, the relative importance of the increase in individual grain weight to compensate for poor stand is less than for the other two yield components.

In most irrigated wheat production systems, common seed rates encountered range from 100 up to 200 kg/ha or more. Research in the Yaqui Valley in Sonora, Mexico, has shown that much lower seed rates can be used if the other aspects of proper wheat crop management are in place, including proper seed distribution, adequate weed, disease and insect control and suitable nutrient and irrigation management. Table 24.4 presents the results for six bread wheat and two durum wheat cultivars planted on the flat in 20 cm row spacing with flood irrigation at 50 and 100 kg/ha seed rates. As can be seen, there was no significant difference between the two seed rates nor was the cultivar by seed rate interaction significant. Table 24.2 has already presented similar data for seven bread wheat cultivars at the same seed rates but planted in beds, which also showed no significant seed rate difference.

Other trials conducted over several years by CIMMYT agronomists have consistently shown that much lower seed rates can be used in most irrigated wheat production systems than are normally used or recommended to farmers. One marked advantage of lower seed rates for wheat is the reduction in crop lodging that normally occurs. This is especially observed when low seed rates are combined with planting wheat on beds. It is common to observe innovative farmers in the Yaqui Valley using wheat seed rates from 40 to 90 kg/ha, yet with bed-planting still maintaining high yields. It seems quite logical that the ability to reduce wheat seed rates with the bed-planting technology can have important, positive implications for the renewed interest in the development and use of hybrid wheat.

TABLE 24.4
Effect of seed rate on grain yield for six bread wheat and two durum wheat cultivars planted on the flat with 20 cm row spacings and with flood irrigation, Yaqui Valley, Sonora, Mexico

^a BW = bread wheat; DW = durum wheat.

^b LSD (0.05) for cultivar means is 642 kg/ha; the seed rate and cultivar by seed rate interaction effects were not significant.

^c Grain yield kg/ha at 12 percent H₂O.

Fertilizer use

Wheat, like all crops, requires adequate, balanced nutrient availability to maximize production potential. Fertilizer use and management is of crucial importance in irrigated production systems since yield potential is high, leading to an extensive removal of essential nutrients. A 7 tonne/ha wheat crop with a harvest index of 40 percent will remove approximately 207 kg N/ha in the above-ground biomass (Hobbs et al., 1998). Given the Nuptake efficiency of irrigated wheat, which normally lies between 30 to 40 percent, the available soil N and applied fertilizer N will need to total about 600 kg N/ha to support the 7 tonne/ha crop. Furthermore, in most situations, because of irrigation availability, usually more than one crop is harvested each year further increasing annual nutrient uptake and required nutrient replacement to maintain or increase system productivity.

The extensive irrigated rice-wheat cropping system in South Asia and China provides a clear example of increasing wheat yields over the past years, increasing fertilizer use and potential sustainability problems. In the state of Punjab in India (most of the area is irrigated), average wheat yields have increased from about 1.8 tonnes/ha in 1970 to almost 4.0 tonnes/ha in 1990 with corresponding increases in N fertilizer use from 54 to 172 kg/ha over the same time period (Hobbs and Morris, 1996).

Similar trends in yield increase and N fertilization have occurred in rice. However, productivity returns to fertilizer use appear to be declining. Estimates for nitrogen efficiency for wheat in India indicate that in 1966 15 kg/ha of wheat were produced from each kg/ha of applied N. In 1992, this had fallen to only 5 kg/ha of wheat for each kg/ha of applied N. During the same period, the ratio for rice declined from 60 kg/ha of rice for each kg/ha of applied N to less than 10 kg/ha of rice for each kg/ha of applied N (Hobbs and Morris, 1996). Estimates such as these, combined with results emerging from long-term trials from several experiment stations in South Asia that are showing downward trending yields over time (more dramatic yield reductions for rice than for wheat), cause concerns about the viability of such intensive, high-yielding production systems if not carefully monitored at the farmer level and properly managed.

Most farmers in northwest India now apply close to the recommended rates of nitrogen and phosphorus, but more frequent examples of responses to and corresponding use of fertilizer sources of other nutrients, such as potassium, zinc and sulphur, are being encountered as more time passes with continuous use of the irrigated rice-wheat rotation.

Nitrogen tends to be the nutrient that is applied at the highest rates and costs the most for most farmers growing wheat under irrigated conditions. In the Yaqui Valley of Sonora, Mexico, an average of 36 percent of variable production costs associated with wheat production is used for fertilizer with a major part spent on N fertilizers (Aquino, 1998). As mentioned earlier, the Nuptake efficiency for irrigated wheat with current farmer practices in the Yaqui Valley ranges from near 30 to 45 percent, averaging around 35 percent (Meisner et al., 1992). CIMMYT agronomists have been researching new N management strategies to improve both Nuptake and Nuse efficiency. Much of this research involves use of the bed-planting of wheat to take advantage of the field access opportunities to allow easy splitting of the N and direct N placement when and where the wheat crop can make the most efficient contribution to crop growth and balanced N demand patterns.

Table 24.5 presents the results of sixteen on-farm trials conducted in the Yaqui Valley testing the effect of both N rate and time of application using the durum wheat cultivar Altar 84, which was planted on beds by all participating farmers. The most common farmer N application strategy in the area is to apply about 75 percent of the total N applied during land preparation and the remaining 25 percent with the first irrigation after crop emergence. This means that most of the N is applied up to a month before the wheat crop is even planted and at least two months before the crop exerts much N demand for growth. The first five treatments in Table 24.5 form an N response curve from 0 to 300 kg N/ha using the common farmer practice of broadcast applying 75 percent of the N pre-plant and applying 25 percent in the water of the first irrigation.

Grain yield and grain protein both increased with increasing N and yellow berry decreased. The last two treatments in Table 24.5 were applied at the rate of 225 kg N/ha, but in one case with 33 percent of the N banded in the bed at planting and 67 percent applied as a banded side-dress at the first node stage. The remaining treatment had all 225 kg N/ha applied as a banded side-dress at the first node stage. The bed system of planting obviously facilitates the band application, especially at the first node. The grain yields and protein contents for the last two treatments (both at 225 kg N/ha) were both significantly higher than for the 225 kg N/ha treatment applied by the prevailing farmer practice and, in fact, were at par with the 300 kg N/ha treatment using the farmer practice. Yellow berry was also less with the last two treatments and similar to the farmer practice of a 300 kg N/ha treatment. The strategy of applying most of the N later clearly improves Nuse efficiency as well as improving quality.

TABLE 24.5
Response of different rates and application timings on grain yield, grain protein content and percent grain yellow berry from sixteen on-farm trials with durum wheat cultivar Altar 84, Yaqui Valley, Sonora, Mexico

^a Common farmer practice for timing N application.

Source: Data courtesy of Dr Ivan Ortiz-Monasterio, CIMMYT wheat agronomist.

TABLE 24.6
Percent wheat grain nitrogen for tillage/residue management and nitrogen management treatments on beds, Yaqui Valley, Sonora, Mexico, 1994

^a Conv. = conventional tillage with new beds formed for each crop; Incor. = residue incorporated; Reduce = permanent beds, reshaped between each crop; Burn = all residue burned after harvest; Partial = removal of wheat straw that was cut by combine (15 to 28 cm stubble left); Remove = residue raked, baled and removed; Retain = all residue chopped and left in the field.

^b Means followed by the same letter do not differ at LSD (0.05); interaction was not significant.

Similarly, Nuse information was collected for the 1994 wheat crop for the tillage/residue trial for which yields are presented in Table 24.3. Grain N contents are shown for a subset of the N management treatments for each tillage/residue management treatment in Table 24.6. As above, when all of the N is banded into the beds at the first node stage instead of a basal application at planting, grain protein contents are significantly higher. When whole wheat plant N removal was measured (Table 24.7), significantly higher N amounts are present with the first node applications, indicating that the wheat plant is taking up more of the applied N and leaving less N to leak into the environment.

TABLE 24.7
Nitrogen removed by the wheat crop for tillage/residue management and nitrogen management treatments on beds, Yaqui Valley, Sonora, Mexico, 1994

^a Conv. = conventional tillage with new beds formed for each crop; Incor. = residue incorporated; Reduce = permanent beds, reshaped between each crop; Burn = all residue burned after harvest; Partial = removal of wheat straw that was cut by combine (15 to 28 cm stubble left); Remove = residue raked, baled and removed; Retain = all residue chopped and left in the field.

^b Means followed by the same letter do not differ at LSD (0.05); interaction was not significant.

Table 24.8 presents the measured N-uptake efficiencies for this trial and clearly indicates the marked improvements in N use. When 150 kg N/ha was applied all as a basal application at planting, uptake efficiency was 35 percent, very near to measurements made in many farmers' fields where most N is applied before or at planting. When the same level of 150 kg N/ha was applied at the first node stage, the uptake efficiency reached nearly 60 percent (a similar but less dramatic trend occurred at the 300 kg N/ha rate). The results point out the potential utility of this strategy to improve the farmer's cost utilization of applied N, as well as a way to reduce N losses into the environment.

TABLE 24.8
Nitrogen-uptake efficiencya for wheat for tillage/residue management and nitrogen management treatments on beds, Yaqui Valley, Sonora, Mexico, 1994

^a [(kg N/ha uptake at given N rate - kg N/ha uptake at 0 N) / N rate] * 100.

^b Conv. = conventional tillage with new beds formed for each crop; Incor. = residue incorporated; Reduce = permanent beds, reshaped between each crop; Burn = all residue burned after harvest; Partial = removal of wheat straw that was cut by combine (15 to 28 cm stubble left); Remove = residue raked, baled and removed; Retain = all residue chopped and left in the field.

^c Means followed by the same letter do not differ at LSD (0.05); interaction was not significant.

Finally, Table 24.9 presents some similar results but includes data concerning direct effects on bread wheat baking quality. A trial was conducted on beds over two years with the bread wheat cultivar Rayon 89, which is widely grown in the Yaqui Valley. In this trial, 150 kg N/ha was applied with three different timings: all at planting, all at the first node stage and one-third at planting and two-thirds at the first node stage (0 N is included for comparison). Yield and grain protein were higher when most or all N was applied at the first node stage. Loaf volumes, however, were markedly improved with the delayed N application treatments, clearly illustrating the potential benefits of this N management strategy for the farmer, the environment and the consumer.

TABLE 24.9
Effect of timing of nitrogen application on yield and quality parameters for the bread wheat cultivar Rayon 89 planted in beds, Yaqui Valley, Sonora, Mexico^a

^a Average of two years results.

OPTIMUM AND SUSTAINABLE IRRIGATED PRODUCTION SYSTEMS

The future of agriculture is characterized by: (i) continued population growth; (ii) a projected requirement for an annual cereal growth rate of 2 to 2.5 percent well into the twenty-first century; and (iii) most arable land suitable for irrigation and with a reliable source of irrigation water is already under production (Hobbs and Morris, 1996). Obviously, to meet the food demand resulting from the combined effects of these factors and given the current, pronounced importance of the irrigated wheat-based production systems in such populous and important developing countries as China, India and Pakistan among others, major increases in cereal production from existing irrigated areas must continue. To accomplish this, an uninterrupted, well-supported breeding effort involving traditional breeding methodologies augmented by new biotechnology applications will be needed to continue the development and release of new ever-higher-yielding cultivars possessing disease and insect resistances along with other important traits.

In concert with this breeding effort is the need to invest more research resources to continue to develop better tillage, nutrient and water management strategies with active farmer participation that are not only sustainable in the context of the environment but that are also practical in terms of farmer confidence and ability for adoption in addition to providing the required, conspicuous increases in food production. A rapid and unambiguous understanding of why productivity appears to be declining in some production systems (most notably those including irrigated, transplanted rice) even though farmers are increasing the levels of production inputs used is critically needed. This understanding is crucial to turna-round the trend where it may be occurring and to prevent it from happening where it may not be happening yet.

In more and more irrigated production areas where wheat is important, there has been the trend in recent years for marked reductions in the diversification of crop rotations. The current prevalence of continuous rice-wheat over large areas in northwest India is a classic example. The economic realities that farmers must face have, unfortunately, brought this to bear. However, data are now being accumulated that strongly indicate that such narrow rotation patterns can exacerbate production declines, especially when continuous cereal crop rotations are followed as compared to more complex rotation systems, particularly those including legume or Brassica spp. crops.

A strong plea is then made to policy makers to support both research efforts and production and marketing policies that can ensure farmers have economically viable, alternate crops that can be included effectively in the crop production systems, thus leading to potential improvements in the productivity base. It may be the catalyst needed to ensure that farmers have access to sustainable as well as optimal irrigated production systems well into the future.

REFERENCES

Aquino, P. 1998. La adopción del método de siembra de trigo en surcos en el Valle de Yaqui, Sonora, México. Informe Especial del Programa de Trigo No. 17b. Mexico, DF, CIMMYT.

CIMMYT. 1996. CIMMYT 1995/96 world wheat facts and trends: understanding global trends in the use of wheat diversity and international flows of wheat genetic resources. Mexico, DF, CIMMYT.

CIMMYT. 1989. 1987-88 CIMMYT world wheat facts and trends. The wheat revolution revisited: recent trends and future challenges. Mexico, DF, CIMMYT.

Hanson, H., Borlaug, N.E. & Anderson, R.G. 1982. Wheat in the third world. Boulder, CO, USA, Westview Press.

Hobbs, P. & Morris, M. 1996. Meeting south Asia's future food requirements from rice-wheat cropping systems: priority issues facing researchers in the post-green revolution era. NRG Paper 96-01. Mexico, DF, CIMMYT.

Hobbs, P.R., Sayre, K.D. & Ortiz-Monasterio, J.I. 1998. Increasing wheat yields sustainably through agronomic means. NRG Paper 98-01. Mexico, DF, CIM-MYT.

Meisner, C.A., Acevedo, E., Flores, D., Sayre, K.D., Ortiz Monasterio, I. & Byerlee, D. 1992. Wheat production and grower practices in the Yaqui Valley, Sonora, Mexico. Special Wheat Report No. 6. Mexico, DF, CIMMYT.

Rajaram, S. & van Ginkel, M. 1996. A guide to the CIMMYT bread wheat section. Wheat Special Report No. 5. Mexico, DF, CIMMYT.