Water, tillage and weed interactions in lowland tropical rice: a review

ELSEVIER Agricultural Water Management 3 1 (1996) 16% 184

Review

Water, tillage and weed

Agricultural water management

article

interactions in lowland .

tropical rice: a review

R.M. Bhagat, S.I. Bhuiyan *, K. Moody

International Rice Research Institute, PO Box 933, Manila, Philippines

Accepted 29 January 1996

Abstract

Interactions among water, tillage and weed management practices are complex, and are further complicated by soil and climatic variabilities and heterogeneities. Studies from the tropical regions on possible effects of tillage and water control on weed emergence and growth in the presence and absence of herbicides have yielded conflicting results due to site specificity. Surface ponding of water in rice (Oryzu satiua L.) reduces weed emergence and growth with variable degrees of success depending upon water depth, nature of weed species and time of ponding. Most studies, however, indicate that shallow ponding for the first few weeks after planting can effectively suppress weeds. Integration of herbicide in a weed control program makes proper water manage- ment more critical. However, good water control is still lacking in most parts of tropical Asia. Investigations of various tillage intensities have revealed that invariably zero tilled soils have more weed population compared with those conventionally tilled in the absence of herbicides. Tillage may bury some weed seeds and expose others that were once deeply buried. Also, repeated tillage will uproot and bury the already germinated weeds. In a diverse weed community situation weed control is effectively achieved if tillage is combined with herbicide application, because tillage is known to enhance herbicide effectiveness. There are reports of identical rice yields being obtained under saturated and flooded water regimes, and zero and conventionally tilled soils. However, effective weed control is required for obtaining such results. Investigations of interactions of tillage intensities and water regimes with weed populations have not been adequately addressed, as most studies have been confined to quantifying competition factors in terms of critical weed population thresholds. Some studies have attempted to explain the nature of competition and its mechanisms. Of the many weed species reported to grow in rice fields, only few actually compete with rice. Generally there are only three to four weed species which are economically important for rice farmers in the tropics. It is these weeds which should be essentially controlled, although

* Corresponding author. Tel: (63-2) 818-1926; fax: (63-2) 817-8470; e-mail: [email protected].

0378-3774/96/$15.00 Copyright 0 1996 Elsevier Science B.V. All rights reserved. PIf SO378-3774(96)01242-5

166 R.M. Bhagut et d/Agricultural Water Munugement 31 (1996) 165-184

total weed control is preferred by most rice farmers. Continuous use of the same control measure in some areas may contribute to a buildup of some tolerant weed species which are difficult to control. Therefore, for effective and sustained weed control, integrated weed management with proper tillage and water control is needed.

Keywords: Tropical weeds; Water-weeds; Water-tillage: Tillage-weeds; Management interactions

1. Introduction

Worldwide, about 148 million ha are planted to rice (Uryza saliva L.) each year, taking into account double and triple cropping. About 90% of this area is in Asia and two thirds in tropical Asia where rice is the most dominant crop grown during the wet season. Dry season rice in tropical Asia is almost entirely grown in irrigated areas.

Enormous amounts of water are generally used in rice production. On average, more than 5000 1 of water are used to produce 1 kg of rice (Bhuiyan et al., 1994). A significant portion of the total water requirement for rice production is used in land preparation. Valera (1976) observed that during the land soaking phase, large amounts of irrigation water are usually lost through cracks, before soil saturation is actually reached; surface drainage also contributes to this loss. This causes delays in completing land preparation, low water use efficiency, and a strained irrigation system (Valera and Wickham, 1978).

Weed control has always been a major factor in rice production throughout mon- soonal Asia, because large amounts of labor have been required in traditional rice culture for weeding. Manual weed control is rapidly being replaced by chemical weed control in many Asian countries (Matsunaka, 1983; Qu, 1989). However, use of herbicides is still limited in countries like Bangladesh, India and Indonesia (Bhuiyan et al., 1994) because of limited availability and high costs (Naylor, 1994). The choice of chemical herbicides depends upon the weed type and the rice culture type (Baltazar and De Datta, 1992). Weed type and degree of infestation in rice fields are often determined by the type of rice culture (irrigated, rainfed lowland, deepwater or tidal wetlands); stand establishment method (transplanted, direct seeded); moisture regime (irrigated, rainfed); land preparation (lowland, upland); and cultural practices (De Datta, 1981; Baltazar and De Datta, 1992). Out of the more than 1000 weed species reported to grow in rice fields, the grass family Poaceae is the most common; Cyperaceae or the sedge family ranks next. Other important weed families are: Alismatuceue, Asteruceae, Fubuceue, Lythruceue and Scrophduriuceue (Smith, 1983). The number of weed species that comprise the major portion of weed flora in any rice field is usually less than ten; in most cases only three or four species are important (Ahmed and Moody, 1982; Kim and Moody, 1980). However, the most dominant weed species are also affected by the water regimes and the type of rice culture (De Datta, 1981; Drost, 1982).

In the Philippines, Thailand and Malaysia, the switch from transplanted to directly seeded rice resulted in major weed shifts, which highlight the need to develop viable and effective weed management practices suitable to the changing situations of small-scale farmers growing rice under various soil and water management practices (Ho and De

R.M. Bhagar et ul./Agricultural Water Management 31 (1996) 165-184 167

Datta, 1988). Such a development would first need a comprehensive study of the existing practices adopted by the farmers in these regions. In temperate east Asia such as Japan, Korea and Taiwan (China), broadleaved weeds and sedge populations are more common than grasses in irrigated rice (Noda, 1973; Kim, 1981). However, in tropical Asia, the moderately warm to high temperature and high humidity favor year round luxuriant growth of almost all weed species (De Datta, 1981). Tillage is a prerequisite for a good seedbed, but it also helps in weed control (Kuipers, 1975; Janiya and Moody, 1983). Soil puddling, which is the common tillage practice in lowland rice cultivation, reduces water loss by percolation and helps to control weeds (Reddy and Hukkeri, 1983). In certain rainfed rice culture, in addition to routine preplant tillage for puddling, farmers carry out additional tillage to destroy the flush of weeds that may occur after rice crop establishment.

Information on changing weed populations in rice due to changing or variable water regimes, use of different tillage methods, herbicide use, and the resulting interaction effects can provide valuable indications of future weed problems and are perhaps indispensable for evolving suitable weed control methods (Fryer and Chancellor, 1970).

In rice culture, tillage, water and weeds are often considered to be closely interlinked. Although their linkages or interactions have been studied in part by many researchers, a comprehensive literature on them is still lacking. The objective of this paper is to analyze and summarize the knowledge on the interactions between water and weeds, tillage and weeds, and water and tillage in tropical rice culture, and to identify common causes for such interactions.

2. Water-weed interaction

2.1. Importance of water for rice and weeds

The amount of water contained in a unit mass or volume of soil and its energy state in the soil are important factors affecting the growth of plants. The amount of water present in soils is highly variable, because of its continuous movement and redistribu- tion. However, in irrigated rice, the variability of soil wetness in the root zone is minimal because a shallow depth of water is generally maintained at the soil surface throughout the crop growth period, which keeps the root zone soil saturated. This practice uses much water. Efforts have been made in the past to save water by either reducing the depth of water on the soil surface (Sivanappan and Saifudeen, 1977; Bhuiyan and Palanisami, 1987) or by keeping the root zone saturated without a water head (Kanwar et al., 1974; Ghani and Rana, 1992). Bhuiyan (1982) reported that rice plants do not suffer from water stress if the soil is saturated and there is no standing water in the field. Also, Tabbal et al. (1992) observed no significant yield difference between rice grown in standing water and that grown under saturated field conditions in 1988-1989 dry seasons; however, yields under saturated soils were lower in 1990- 1991 dry seasons because of more weed growth compared with the previous dry seasons.

Submergence and saturated conditions markedly reduce soil redox potential, indicat- ing enhanced reduction of the soil (Ponnamperuma, 1972; Naphade and Ghildyal, 1974).

168 R.M. Bhqut et crl./Ap-iculturul Wuter Muncrgrmmt 31 (1996) 165-184

Reduced soil conditions increase the availability and uptake of nutrients to rice plants, especially P and Mn, which may lead to higher rice yields under irrigated conditions (Obermueller and Mikkelson, 1974; Bhatnagar and Sharma, 19751, but at the same time availability of Fe is also increased which sometimes results in toxic effects (Ponnam- peruma, 1972). Reduced soil conditions help in decomposing uprooted weeds and hamper weed seed germination (De Datta et al., 1970; Mercado, 1979; Moody, 1983).

It is widely recognized that rice grown under submerged soil conditions competes better with weeds than in dryland conditions. Submergence of rice fields is thus an integral part of traditional weed control (Matsunaka, 1983). Submergence hinders weed germination and suppresses the weed population of already germinated weeds, depend- ing upon the nature of weed flora. Williams et al. (1990) compared the growth of several weeds in water seeded rice under shallow (5 cm), moderate (10 cm> and deeper (20 cm) continuous floods. Without herbicide, 20 cm water gave better weed control than other water depths. However, with herbicide, weed control improved at all water depths, but weeds were not fully controlled in shallow water even with herbicide. This indicates that herbicide effectiveness is linked with water depth. The nature of the weed flora and selection of the herbicide is another factor limiting weed control, as sometimes desirable weed control is not obtained in spite of herbicide application, because of the wrong choice of the herbicide.

2.2. Water management and weed emergence

Most weeds propagate from seeds; only a few propagate vegetatively. Agricultural soils generally contain a large reservoir of weed seeds (Vega and Sierra, 1970; Zimdahl et al., 1988). The number and type of weed seeds present in the soil reservoir are largely determined by the water holding capacity and pH of the soil, past weed control practices, and tillage and land preparation practices. Too low or too high soil water content may also restrict weed seed germination (Zimdahl et al., 1988). The amount of water also influences the periodicity in the germination of weeds (Sahu, 1978; Diop and Moody, 1984). Because many weeds cannot germinate under flooded conditions, excessive water in the soil serves as an effective means of weed control (Moody, 1978). Arai et al. (1955) reported that the amount of weed emergence in submerged plots was about 30% of that in saturated plots (moisture content of 80-90%) and as low as about 17% of that in upland plots (moisture content of 40-60%). Maintainance of a few centimeters of water over the soil surface can thus be used to suppress weed emergence in rice soils.

Moist or saturated soils favor the emergence of grasses and sedges. Once established, these weeds are difficult to control by flooding. However, flooded conditions during and after transplanting or wet seeding (broadcasting of sprouted seeds) suppress grasses, but encourage sedges to dominate (De Datta, 1981; Mabbayad et al., 1983). The time of flooding after rice seeding also affects weed emergence. Echinochloa glabrescens Munro ex Hook. f. was found to dominate when flooding of the rice field was delayed for 5 to 10 days after seeding (Janiya, 1984). Also, delay in flooding for up to 12 days increased the density and dry weight of E. glabrescens and significantly decreased the yield of wet-seeded rice (Drost and Moody, 1982). Generally, delay in flooding after

R.M. Bhagat et al./Agricultural Water Management 31 (1996) 165-184 169

planting encourages weed emergence because most weeds cannot germinate under water. Hence flooding of the field immediately after planting can be an effective way to suppress weed emergence.

2.3. Water management and weed growth

Growing transplanted rice under submerged conditions is perhaps the first traditional step towards weed control (Matsunaka, 1983; Moody, 1983). The composition and growth of rice weed communities is strongly influenced by water management practices (Ahmed and Moody, 1982). Studies in Asia also revealed that weed growth suppression can be achieved through water control (Kim, 1980; Pablico and Moody, 1982; Estomi- nos and Moody, 1983; Nartsomboon and Moody, 1988a,b). Moturi (1977) reported that weed density and dry weight were higher under saturated conditions than under submergence. Continuous submergence (10 cm) resulted in lower density (63 plants m-*) and weed weight (17.5 g m-‘) than continuous saturation which had a weed density of 86 plants m-’ and weed weight of 19 g m- 2 (Misra et al., 1981). Tabbal et al. (1992) reported that by maintaining continuous shallow submergence, especially during the vegetative period, weed growth was effectively suppressed. Exposure of the soil surface at any time during the first few days after transplanting encourages weed germination. Zimdahl et al. (1987) reported that three lowland weeds (Cyperus difsormis L., Fimbristylis miliacea (L.) Vahl, Leptochloa chinensis (L.) Nees) grew poorly at soil water levels below field capacity and under flooded conditions compared with that at 120% and 160% of field capacity. Weed growth in rice fields is therefore strongly influenced by the amount of water present, as even short periods of dry soil can greatly encourage weed growth.

2.4. Water management and herbicide effectiveness

Water management plays a major role in any weed control program regardless of whether herbicide is used or not. The presence or absence of water during rice stand establishment determines, in a large part, the mixture of weed flora (Hill et al., 1994). Water depth can be used to control many weeds, but there are many other weed species that are relatively unaffected by water depth. When herbicides are integrated into the weed control program, water management becomes especially critical. Water levels must be manipulated to enhance exposure of the weed foliage to some herbicides or to cover weed foliage for other herbicides; it is equally important to prevent the movement of herbicide-treated water from the fields as to prevent pollution of the environment (Bayer and Hill, 1989).

Herbicide activity in soils and effectiveness against weeds is related to field water status (Floresca et al., 1970; Bhan and Singh, 1979: Ho et al., 1992). In Asia, several herbicides are now used for weed control in transplanted rice, due largely to rising labor costs (De Datta and Flinn, 1986). The phenoxys 2,4-D and MCPA, are effective against sedges and broadleaved weeds when applied postemergence (15-25 days after trans- planting). Butachlor and thiobencarb, which control most annual grasses, broadleaved weeds and sedges, are applied preemergence or early postemergence to weeds (Baltazar

170 R.M. Bhagat et al./Agricultural Water Management 31 (1996) 165-184

and De Datta, 1992). Propanil is used for postemergence control of annual grasses, and some broadleaved weeds and sedges. Preemergence herbicides that effectively control weeds in wet seeded rice in the Philippines include butachlor and thiobencarb (Moody, 1985). Molinate has been proven effective against Echinochloa crus-galli (L.) P. Beauv. ssp. hispidula (Retz.) Honda and Echinochloa colona CL.1 Link, while pretilachlor was excellent for controlling Echinochlou spp., and Leptochlou chinensis (L.) Nees in Malaysia (Ho and Md. Zuki, 1988). The importance of appropriate selection of herbicide for target weeds cannot be overemphasized.

Many preemergence herbicides for control of grassy weeds have certain prerequisites in order to realize their full potential for weed control. The field must be flooded when the herbicide is applied. The weeds must be at least two-thirds submerged and the condition maintained for at least a week (Ng, 1986). On the other hand, postemergence herbicides such as propanil require fully drained fields to expose the grassy weeds for maximum contact with the herbicides. A mixture of propanil and molinate requires the rice fields to be flooded within 48 h to enable the molinate residue to control germinating grasses (Soong, 198 1). Herbicide use in transplanted rice in tropical Asia is often limited by the need for good water control, because sometimes fields have to be drained before and reflooded after application, a degree of water control not generally available in tropical Asia (De Datta, 1980).

Herbicide behavior in rice or weeds and in soil and water has been less investigated than herbicide efficacy evaluation under irrigated transplanted conditions. Some studies have specifically investigated the effect of soil, water and environmental factors on herbicide efficacy (Kawamura and Hirai, 1975; Rahman et al., 1977; Jikihara et al., 1981; Pathak et al., 1989). Although most of the studies show that herbicide activity and rice susceptibility to herbicide are greater in flooded than nonflooded conditions, it is not fully understood how flooding enhances herbicide activity (Baltazar and De Datta, 1992). However, many studies on herbicide fate in soil and water indicate that rice herbicides do not persist in the environment (Ishikawa et al., 1975; Chen, 1981; Patcharin and Prateep, 1988). The risk of herbicide contaminating groundwater and polluting the environment, therefore seems minimal. However, herbicide integeration with other control measures, which can reduce the amount of herbicide aplication and thus reduce herbicide residue accumulation in the environment, if any, should be further studied.

2.5. Water management and weed species

Poor water control often contributes to increases in the different weed species (Navarez et al., 1979). Cabailo (1925) reported that there were 14 families, 73 species and 417 individual weeds under upland conditions and eight families, 46 species and 397 individuals under lowland conditions. The continuous presence of water during vegetative growth formed an effective means of weed control. Only a few families, which were either amphibious or aquatic in nature, were able to grow under such conditions.

Weed species respond differently to changing water regimes (Tauro, 1970; Janiya and Moody, 1982). According to Arai et al. (19551, the distribution of a weed species and its

R.M. Bhagur et al./Agricultural Water Management 31 (1996) 165-184 171

growth varies according to soil moisture conditions. Tanaka (1976) reported that in flooded conditions, C-3 weeds dominated over C-4 weeds. In water saturated (at SO-90% of maximum soil moisture capacity) and upland plots (4060% of maximum moisture capacity), C-4 species accounted for more than 90% of the total dry weight, in sharp contrast to 10% in submerged plots (6 cm standing water). The speed of flooding and intermittant draining and drying of the soil also affects the weed species, particularly some algae and aquatic weeds (Hill et al., 1994). Water depth impacts the species mixture by suppressing such weeds as Echinochloa and Lepfochlou spp. while having less effect on others (Raju and Reddy, 1987).

Soil moisture status after planting is the major factor influencing the composition of weed flora (Drost and Moody, 1982). For example, the dominance of Echinochloa species was favored by saturated conditions, while Leptochloa chinensis (L.) Nees emerged best when the soil moisture was just below saturation. In rice fields, change in water status from irrigated to rainfed would shift the weed population from semi-aquatic and aquatic to terrestrial species. For example, unweeded fields continuously prepared wet had consistently higher Scirpus maritimus L. population than those continuously prepared dry (Bernasor and De Datta, 1986). Continuous dryland preparation also resulted in a complete weed shift from lowland to upland annual species.

Weed species grow differently at different moisture levels. E. Crus-galli grows best at soil moisture of 80% of water holding capacity. However, its emergence and growth were reduced with increased submergence up to 15 cm (Bhan, 1981). Also, Smith and Fox (1973) observed that flooding up to 5 cm controlled E. crus-galli, but at field capacity or saturation, satisfactory stands were developed. On the contrary, Lo (1988) observed that pregerminated seeds of Echinochlou species were more tolerant to submergence. Also, seeds of some aquatic weeds germinate readily under water. Once germinated, these weeds continue to grow even after flooding takes place.

In addition to degree of submergence, the composition of weed flora is often strongly influenced by the landscape position. The higher water ponding potential in coastal areas having a higher water table favored the growth of aquatic weeds such as Zpomea aquatica Forssk., Nymphoides indica (L.) O.K., Blyxa malayana Rich., Hydrilla uerticillutu (L.f.) Royle, Otteliu alismoides (L.) Vahl whereas the inland areas with lower ponding potential and low water table favored Cyperus iris L., Cyperus babaken- sis Steud., Scirpus supinus L. and Melochiu corchorifolia L. (Soo, 1972).

Rapid change in soil and water management practices in Malaysian rice farms in the Muda irrigation project has caused a shift in weed populations from annuals to perennials, from shallow emerging to deep emerging weeds and from less competitive to more competitive weeds (Noda, 1973). To overcome this problem, integrated weed management, which emphasizes a combination of control methods rather than a single method, has a far greater chance of success. However, the shift in weed population from the more difficult to control perenniels to the easier to control annual weeds due to change in soil and water management, may be an advantage in a total management scheme (De Datta, 1988). The shifts in weed population over time are basically a consequence of changes in rice establishment methods. The new rice establishment techniques now adopted should have a component of weed control, so that if there are any weed shifts, those can be effectively controlled.

172 R.M. Bhugat et d/Agricultural Wutrr Management 31 (1996) 165-184

3. Tillage-weed interaction

3.1. Tillage needs in rice culture

Weed control is one of the main reasons for tilling rice soils. Other reasons include good crop establishment, incorporation of fertilizer and crop residue into the soil, and in the case of wetland rice culture, reduction of water permeability of soil. Puddling is the most common tillage practice to achieve this objective (Sharma and Bhagat, 1993). The degree of reduction in permeability due to puddling varies widely with soil type and its structural development (Sanchez, 1973; Lal, 1985a). De Datta and Kerim (1974) have reported that in Mahaas clay, percolation was considerably higher in a non-puddled soil than a puddled soil. Well-structured clay soils respond more to puddling than soils with coarse to medium texture and having single grain or poorly developed structures (Lal, 1985b). Puddling buries weeds in the lower layers of the slurry, where they decompose by anaerobic action to form ammonium compounds which are retained much better than nitrate in the soil and can be used directly by the crop. Puddling, therefore, not only destroys the weeds but also converts them into a very useful form of fertilizer (Wrigley, 1969).

Puddling is preceded by tilling operations which, for rice, normally consist of one or two ploughings and harrowings. Barker (1970) and Manuel et al. (1979) found that the number of harrowings given by farmers was correlated with weed growth in transplanted rice.

Tillage in rice culture is intended to destroy existing weeds and to stimulate the germination of weed seeds in the top soil layers so that resulting weed seedlings can be killed by disking or harrowing before the crop is planted. Different kinds of tillage have been evaluated for rice-weed competition. Many studies have indicated that conventional tillage is superior to zero and minimum tillage in weed control and grain yield (Janiya and Moody, 1983; Shad and De Datta, 1986). Weed control requirement was found to be maximum in the zero-tillage treatment and next highest in the treatment with one ploughing. However, one ploughing with a power tiller or two ploughings with a country plough combined with one harrowing, reduced the total weed weight, although the rice yield was not affected under different land preparation methods (Ahmed, 1982). Where primary cultivation is eliminated or reduced, such as in zero or minimum tillage, changes in weed flora may take place, requiring increased use of manual or chemical weed control. When perennial weeds such as Paspalum distichum L. are present in the field, increased tillage is essential to minimize their growth (De Datta and Bernasor, 1988). Deep tillage (up to 25-30 cm) has been shown to reduce annual and perennial weeds in transplanted rice (Sarkar and Ghosh, 1979). Deep ploughing will bury weed seeds and some seeds loose their vitality when buried deep in the soil. However, Satyal (1978) reported that depth of tillage had no effect on weed growth in transplanted rice.

3.2. Tillage and weed emergence

Tillage serves only as a temporary means of weed control as the soil contains many ungerminated weed seeds. Plowing may bury weed seeds at a depth that prevents

R.M. Bhagat et d/Agricultural Water Management 31 (1996) 165-184 173

germination but may also expose other, once deeply buried weed seeds to conditions conducive to germination (Stoskopf, 1985). The specific effects of tillage on weed seed germination have not been thoroughly investigated for lowland rice systems. However, it is widely accepted that tillage influences weed germination through deeper seed burial and inducing dormancy (Zimdahl et al., 1988). Smaller seeds with long dormancy survive longer with deeper burial, but their emergence rate is lowered (Chepil, 1946). Watanabe and Hirokawa (1975) reported that shallow tillage (5 cm) every month changed the magnitude of weed species emergence but not the pattern of emergence. However, Zimdahl et al. (1988) observed that tillage caused weed emergence regardless of the time that it occurred. They concluded that stimulation of germination and emergence occurred immediately after soil tillage. In their experiment, 58% of total emergence occurred within 6 weeks after tillage; only 0.7% of the weeds emerged within 2 weeks of tillage. In addition to tillage, the delay in emergence was related to the flooded lowland environment. Through improvement of air exchange in the soil, tillage stimulates weed seed germination, but it can also uproot and eliminate already germinated weeds. Therefore time and intensity of tillage are important factors govem- ing the weed population in rice culture.

3.3. Tillage and weed species

Tillage influences the build up or suppression of some weed species, but there are very few studies on the effects of tillage intensity on weed species shift in rice culture. Intensive tillage may reduce Echinochloa infestations, which is among the most severe weed species causing problems in rice (Lubigan and Vega, 1971; Noda, 1973; Smith, 1983; Rao and Moody, 1987), but low intensity tillage may increase its incidence. Researchers have found that low intensity tillage and inadequate chemical weed control increased P. distichum incidence in rice fields in India (Seth et al., 1971); Pistia stratiores L. in the Philippines (Buangam and Mercado, 1977); and Eleocharis kuroguwai Ohwi in Japan (De Datta, 1981). Field studies on the growth of Echinochlou stugnina (Retz) P. Beauv., a perennial grass weed in the Muda irrigation project area of Malaysia, revealed that rice fields which were given only one dry rotovation were most severely infested. However, when the fields were rotovated twice under inundated conditions and thoroughly puddled, the sprouting capacity of E. stagnina was drastically reduced (Ho et al., 1992). In addition to soil puddling, regular weeding along the dikes also controlled E. stugnina infestation (Ho and Itoh, 1990). Another study on Mursilea minutu L., a perennial fern, indicated that land levelling under wet-field conditions aggravated the M. minutu problem by dispersing the rhizomes over the field, while dry plowing and dry land levelling exposed the rhizomes to the hot sun and reduced weed infestation. In the wet season, rotovation followed by two rakings resulted in lower M. minutu density than one raking (Ho, 1985).

Continued puddling of rice fields results in fewer weed species (Ahmed, 1979; Mukhopadhyay, 1981), but a higher proportion of broadleaved weeds in the weed flora (Moody, 1982). Harwood and Bantilan (1974) have indicated that Eleusine indicu (L.) Gaertn., Digitaria sanguinalis (L.) Stop., Amaranthus spinosus L. and Cyperus rotun- dus L. are most sensitive to puddling, Echinochloa colona (L.) Link and Portulaca

174 R.M. Bhagat et al./Agricultural Water Management 31 (1996) 165-184

olerucea L. are moderately sensitive and Cyperu~ iris L. is least sensitive. Repeated cultivation at l-3 week intervals before seeding has also been shown to reduce E. Crus-galli and other annual grass infestation (Sarkar and Ghosh, 1979).

Puddling is an energy intensive process (Sharma et al., 1988). To reduce energy used in puddling, different degrees of tillage have been tried (Sharma and De Datta, 1986; Bhagat et al., 1994), which have produced varying effects on weed density (Bemasor and De Datta, 198 1; Janiya and Moody, 1983). Janiya and Moody (1983) reported that the major weeds in minimum and maximum tillage rice plots were Monochoria vaginalis (Burm. f.) Presl, E. crus-gulli ssp. hispidulu, E. glubrescens, L. chinensis, and Cyperus difformis L. In their experiment, P. distichum, the only perennial weed, L. chinensis and E. crus-gulli ssp. hispidulu were dominant in the zero tilled plots. P. distichum was present in the minimum tillage plots in the second rice crop.

3.4. Tilluge and herbicide effectiveness

When herbicides were used, yields from the zero tillage plots were significantly lower than those from the other tillage treatments, which did not differ from each other, implying that efficiency of herbicides in controlling weeds varies with degree of tillage (Janiya and Moody, 1983). However, Bemasor and De Datta (1988) observed that irrespective of tillage level, herbicide application significantly increased P. distichum density in an experiment on Maahas clay soil in the Philippines. Proper tillage suited to a particular soil may thus be essential for the effectiveness of a herbicide. De Datta et al. (1979) reported that after two crops without tillage, perennial weeds became dominant because they could not be controlled by the pre-plant herbicide used. Usefulness of zero or reduced tillage is therefore sometimes questionable and may be limited to areas where perennial weeds are not present (Moody and De Datta, 1980). Reduced tillage also requires more herbicide use compared with conventional tillage to control weeds (Moody and De Datta, 1980). Pablico and Moody (1982) found that the efficiency of herbicides like 2,4-D in controlling weeds is directly linked to the degree of tillage. When 2,4-D was applied, weed weights in plots which were harrowed once and in the zero tillage plots were not significantly different from those in the unweeded plots. However, in the plots with conventional tillage, 2,4-D effectively controlled weeds. Janiya and Moody (1983) and Bemasor and De Datta (1988) observed that continuous herbicide application in rice did not decrease the weed population, instead the weed population increased in the later years. Because neither tillage nor herbicide use alone provides total control of a diverse and persistent weed community, it is important to combine both methods to achieve effective and sustained weed control and to obtain high and stable grain yields.

4. Water-tillage interaction

4.1. Water, tilluge and soil structure

Puddling, which is considered to be a pre-requisite for modem rice culture in Asia, is accomplished by a series of tillage operations, usually by maintaining a depth of water

R.M. Bhagat et al./Agricultural Water Management 31 (1996) 165-184 175

over the saturated soil (Salokhe and Shirin, 1992). Puddling, accompanied by a destruction of soil aggregates, results in a decrease in the volume of transmission pores. After rice harvest, the puddled soil remains wet and reduced at least for a few days, and when dried becomes hard and has a high bulk density (Sanchez, 1976). Such soil upon tillage forms large clods, which vary in size depending upon the clay content and the implement used for tillage (Bhushan et al., 1973; Naravani and Sirohi, 1978). Puddling destroys soil structure, which in turn influences both water holding and water transmit- ting capacity of the soil. Important aspects of soil structure are stability and size distribution of aggregates and sometimes roughness brought about by tillage (Cresswell et al., 1991). Antecedent soil water content has important effects on soil structure produced by tillage operations (Bhushan and Ghildyal, 1972; Hoyle et al., 1972; Johnson et al., 1979; Adem et al., 1984; Tisdall and Adem, 1986). Johnson et al. (1979) observed that moldboard ploughing of a wet silt loam to silty clay loam soil increased surface roughness and that clods resulting from ploughing wet soil had lower wet stability than clods formed from ploughing at soil water content near the lower plastic limit, which markedly influenced soil water behavior. The relative abundance of a particular size range of aggregates and proportion influences the soil water intake. Aggregates < 0.25 mm in diameter and individual soil particles play a major role in decreasing infiltration rates through clogging of pores and surface seal development (De Ploey and Poesen, 1985; McCaig, 1985).

4.2. Water, tillage and soil porosity

Tillage results in changes in soil porosity and pore size distribution which dictate soil water retention characteristics and influence soil bulk density. Puddling by country plough and harrow or by rototiller decreases total soil porosity only slightly, but markedly changes porosity distribution, with both storage and residual porosity increas- ing at the expense of transmission porosity (Painuli et al., 1988). Decreasing bulk density increases total porosity, and hence, increases the amount of water held at high soil water potential and decreases the amount held at lower potential (Unger, 1975; Klute, 1982). However, the response may depend on soil texture. For example, Unger (1975) observed that disturbing the natural soil structure decreased the water retention of coarse textured soils and increased the retention of fine textured soils relative to that of natural soil cores at a manic potential of -0.033 MPa. At a manic potential of - 1.5 MPa, disturbed soil samples of all textures retained slightly more water than undisturbed samples, but the percentage change was greater for coarse than for fine-textured soils. This indicates that soil texture plays an important role in soil water retention following soil disturbance. Besides the textural response, tillage may further influence water retention if it incorporates crop residues or if it alters the distribution of sand, silt and clay in the soil by mixing particles from different soil horizons (Miller and Aarstad, 1972; Bhagat, 1990). Bhagat (1990) observed that in a coarse textured soil incorporation of residues significantly increased water retention capacity.

4.3. Water, tillage and soil hydraulic properties

Tillage operations markedly affect soil hydraulic properties including both saturated and unsaturated hydraulic conductivity, soil water content, soil water retention and soil

176 R.M. Bhagat et al./Agriculturul Water h4unqement 31 (1996) 165-184

water diffusivity (Klute, 1982). These properties in turn affect hydraulic processes such as infiltration, water flow through soils, drainage, and evaporation.

Surface as well as subsurface soil conditions influence water infiltration. Tillage alters both surface and subsurface conditions, through the shearing action of the tillage tool and may determine the amount of water required for land preparation. Puddling causes destruction of soil structure in rice, which results in a decrease in soil water transmission - a condition often desired for rice. Consequently, puddling decreases percolation and therefore better retains standing water in the field (Sharma and De Datta, 19861, which may also decrease the irrigation requirement.

4.4. Water, tillage and evaporation

A significant portion of water present in the soil is lost to the atmosphere through evaporation. Tillage exposes the subsoil to the atmosphere, which increases the rate of water loss from the soil through evaporation. By loosening the soil near the surface, tillage may reduce capillary water rise to the surface and thereby decrease evaporative losses of deeply stored soil water. However, the prevention or the substantive reduction of evaporative losses of water is mostly not cost-effective in flooded lowland rice fields.

4.5. Water, tillage and soil texture

Rice response to tillage varies with soil texture and climatic water balance (Sharma and De Datta, 1994). Depending on the soil texture, tillage may induce a gain or loss in soil permeability which may affect rice yield through better retention of surface water. Sharma et al. (1987) have shown that the grain yield of rice was significantly effected by tillage in a sandy loam, but not in a clay loam soil, with a shallow water table. Soil compaction has been suggested as an alternative to puddling in coarse to medium textured soils (Ghildyal, 1978) because their physical properties change little with puddling (Lal, 1985a; Lal, 1985b; Sharma and De Datta, 1986). Puddling as well as compaction decreases the water permeability of soils by decreasing the volume of water transmission pores (Sharma and De Datta, 1986). The reduction, however, depends on the intensity of the treatment and percentage of silt and clay present in the soil. Sharma and Bhagat (1993) observed that in soils with < 70% sand, puddling as well as compaction are equally effective in decreasing water percolation to satisfactory levels for growing rice. The choice between the two depends upon factors like susceptibility of rice to compaction levels, residual effects of puddling and compaction on upland crops grown after rice, and regeneration of soil structure after the puddled crop. However, in soils having > 70% sand, compaction rather than puddling is effective in decreasing water permeability.

Puddling in coarse textured soils induces high bulk density and low permeability in subsurface layers (Sur et al., 1981; Prihar et al., 1985b), the effect of which continues through the post-rice season. The soil profile below the puddled zone remains unsatu- rated and has high impedance to root growth in post-rice crops. As water and nutrient availability depend primarily on the soil volume exploited by root systems (Drew, 1978), unfavorable conditions in the soil profile may severely limit crop yields through

R.M. Bhagat et al./Agricultural Wuter Management 31 (1996) 165-184 177

adverse effects on root growth, proliferation and activity (Unger, 1979). Meelu et al. (1979) reported a decrease of 0.3-0.4 Mg ha -I in wheat (Triticum aestiuum L. amend Thell.) yield following rice than following maize (Zea muys L.) in a sandy loam soil. This indicates that tillage has a detrimental effect on soil properties, which adversely affects the post rice crops. Management options for sustained production in these soils include maintaining a wetter and more fertile rhizosphere and/or inducing deeper and denser rooting (Gajri et al., 1992). Deep tillage is one option to shatter the subsurface dense layers formed in a rice-upland cropping system, which would help in proliferation and penetration of roots of post rice crops to deeper layers, thus exploiting more water and nutrients (Bhagat and Acharya, 1987).

4.6. Water, tillage and energy input

One of the requirements of an efficient rice based production system is to decrease energy inputs. Tillage, irrigation and fertilizer constitute the major energy inputs in a crop production system. As energy costs continue to increase, it becomes imperative to develop energy efficient crop production systems. At low yield levels, water and nutrients can substitute for each other (Prihar et al., 1985a; Prihar et al., 1989) and tillage increases efficiency of water and fertilizer use (Power, 1983; Sharma, 1985; Eck and Unger, 1985). Interactions among energy intensive inputs of tillage, irrigation and fertilizer-N can be gainfully exploited to combat soil and management related stresses for improved crop performances. Gajri et al. (1992) reported that tillage reduced soil strength in the tilled zone, caused deeper and denser rooting, increased both the E (evaporation) and T (transpiration) fraction of ET and also the efficiencies of N and water use by the crop. Energy input in conventional tillage was 280 kwh more than that in a no-till system; however, the former system produced more grain per unit of energy consumed. The increase in N and water use efficiency occurred because energy input in tillage operations in conventional tillage was far less than that in the extra N and irrigation required in the no-till system to achieve a given yield.

5. Future research challenges

Efficient water management, proper tillage and judicious use of herbicides are needed for sustainable rice production. Much work has been reported on water management practices of rice in combination with other factors like tillage, weed management and nutrient management practices. However, research gaps such as the following still exist and need attention:

1. There is concern over excessive chemical use in rice production. Weed control measures using limited and environment-friendly chemicals are needed. The potential hazard of the chemicals may be reduced using integrated chemical and non chemical control measures, for which further strategic research is required.

2. The inherent capacity of non chemical weed control means such as water and tillage, needs further investigation, keeping in view the need for achieving high water use efficiency and low energy consumption in tillage.

178 R.M. Bhagat et al./Agricultural Water Management 31 (1996) 165-184

3. The interactive effects of water-weeds, tillage-weeds, and water-tillage should be quantified for determining their optimal input levels under major rice based cropping systems.

4. In-depth, comprehensive studies to develop weed management strategies for greater ecological and economically sustainable crop production systems should be carried out. Issues of weed seed dormancy, germination, and emergence behavior under varied tillage and water management practices should also be addressed.

5. In future, with the increase in the cost of farm labor, direct seeding of rice will become increasingly popular replacing the transplanted rice system. The ecological changes in the weed regime induced by such changes in the rice crop establishment method need to be thoroughly studied in order to develop sustainable weed management strategies.

Most research conducted in the past has not adequately addressed many of the issues using a systems approach, which often requires a focus on the entire cropping sequence for long enough time. This must be kept in perspective in formulating future research plans on the above issues.

References

Adem, H.H., Tisdall, J.M. and Willoughby, P., 1984. Tillage management changes, size distribution of aggregates and microstructure of soil used for irrigated row crops. Soil Tillage Res., 4: 46 l-476.

Ahmed, N.U., 1979. Weeds in cropping systems as affected by hydrology and weeding regime with emphasis on dry seeded rice. MS. Thesis, University of Philippines, Los Banos, Laguna, 241 pp.

Ahmed, N.U., 1982. Past research in weeds and weed control in Bangladesh and future directions. Report, Workshop on Cropping Systems Research in Asia, International Rice Research Institute, Los Banos, pp. 539-547.

Ahmed, N.U. and Moody, K., 1982. Weeds in rainfed rice cropping systems as affected by landscape position. Proceedings International Conference on Plant Protection in the Tropics, Kuala Lumpur, pp. l-7.

Arai, M., Miyahara, M. and Yokomori, H., 1955. Ecological studies of weeds on arable land. IV. On the types produced by the adaption of soil water to weeds (in Japanese, English abstract). J. Kanto-Tosan Agric. Exp. Sta., 8: 52-62.

Baltazar, A.M. and De Datta, SK., 1992. Weed management in rice. Weed Abstr., 41: 495-508. Barker, R., 1970. The economics of rice production. In: Rice Production Manual. University of Philippines at

Los BaZos, Laguna, pp. 286-305. Bayer, D.E. and Hill, J.E., 1989. Weed control practices and problems in direct-seeded rice culture. In: Weed

Problems and their Economic Management. Asian-Pacific Weed Science Society and Korean Society of Weed Science, pp. 53-56.

Bemasor, P.C. and De Datta, S.K., 1988. Long term effects of tillage, cultivar, and herbicide on weed shift and control in broadcast-seeded flooded rice. Soil Tillage Res., 12: 197-2 12.

Bemasor, P.C. and De Datta, S.K., 1986. Chemical and cultural control of bulrush (Scirpus maritimus L.) and annual weeds in lowland rice (Oryza satiua). Weed Res., 26: 233-244.

Bemasor, P.C. and De Datta, S.K., 1981. Long term effects of reduced tillage on weed shift in lowland rice. Paper presented at the 12th annual conference of the Pest Control council of the Philippines. University of Philippines at Los BaZos, Laguna, 15 pp.

Bhagat, R.M., Sharma, P.K. and Verma, T.S., 1994. Tillage and residue management effects on soil physical properties and rice yield in north western Himalayan soils. Soil Tillage Res., 29: 3233334.

Bhagat, R.M., 1990. Effect of tillage and residue management on hydro-thermal regime, nutrient uptake and yield of winter wheat in a river deposit. Soil Tillage Res., 17: 315-326.

R.M. Bhagat et al./Agricultural Water Management 31 (1996) 165-184 179

Bhagat, R.M. and Acharya, CL., 1987. Effect of soil management on rainfed wheat in Northern India. I. Hydro-thermal regime and root growth. Soil Tillage Res., 9: 65-77.

Bhan, V.M., 1981. Effects of hydrology, soil moisture regime and fertilizer management on weed population and their control in rice. In: Weed Control in Rice. International Rice Research Institute, Los Banes, pp. 47-56.

Bhan, V.M. and Singh, D., 1979. Tips in weed control in rice. Indian Farmers’ Digest, 12: 1 l- 12. Bhamagar, O.K. and Sharma, S.N., 1975. Comparative study of different methods of seeding on the yield of

rice with and without weeding. Indian J. Agron., 20: 58-59. Bhuiyan, S.I., Sattar, M.A. and TabbaI, D.F., 1994. Wet seeded rice: water use efficiency, productivity and

constraints to wider adoption. Paper presented at the International Workshop on Constraints, Opportunities and Innovations for Wet Seeded Rice, Bangkok, May 31-June 3 1994, 19 pp.

Bhuiyan, S.I. and Palanisami, K., 1987. Increasing efficiency of water use on irrigated rice farms. Paper presented at the International Rice Research Conference, Hangzhou, 21-25 September 1987.

Bhuiyan, S.I., 1982. Irrigation system management research and selected methodological issues. IRRI Research Paper Series 8 1, International Rice Research Institute, Los Banos.

Bhushan, L.S. and Ghildyal, B.P., 1972. Influence of radius of curvature of moldboard on soil structure. Indian J. Agric. Sci., 42: l-5.

Bhushan, L.S., Varade, S.B. and Gupta, C.P., 1973. Influence of tillage practices on clod size, porosity and water retention. Indian J. Agric. Eng., 43: 466-471.

Buangam, T. and Mercado, B.L., 1977. The life cycle of water lettuce (Pista sfratiotes L.). Philipp. Weed Sci. Bull., 2 (2): 11-15.

Cabailo, B.C., 1925. Weeds in the rice fields and their effect on the yield of grain. Philipp. Agric., 14: 359-371.

Chen, Y.L., 198 1. Degradation of butachlor in paddy fields. Proceedings 1980 Conference on Weeds and their Control in Asia, Taipei, pp. 121- 142.

Chepil, W.S., 1946. Germination of weed seeds.11. The influence of tillage treatments on germination. Sci. Agric., 26: 347-357.

Cresswell, H.P., Painter, D.J. and Cameron, K.C., 1991. Tillage and water content effects on surface soil physical properties. Soil Tillage Res., 21: 67-83.

De Dana, S.K., 1980. Weed control in rice in South and Southeast Asia. Extension Bulletin no. 156. Published by the Asian-Pacific Food and Fertilizer Technology Center, Taipei, pp. 24.

De Datta, SK., 1981. Weeds and weed control in rice. In: Principles and Practices of Rice Production. John Wiley, NY, pp. 460-5 12.

De Datta, S.K., 1988. Progress in weed science and control technology in tropical rice. Proceedings 1988 Tropical Weed Science Society Conference, Thailand, pp. 14-35.

De Datta, S.K. and Bemasor, P.C., 1988. Agronomic principles and practices of rice ratooning. In: Rice Ratooning. IRRI, Los Banos, pp. 163- 176.

De Datta, SK. and Flinn, J.C., 1986. Technology and economics of weed control in broadcast seeded flooded tropical rice. Proceedings 10th Asian Pacific Weed Science Society Conference, Chiangmai, pp. 51-74.

De Datta, S.K. and Kerim, M.S.A.A.A., 1974. Water and nitrogen economy of rainfed rice as affected by soil puddling. Soil Sci. Sot. Am. Proc., 38: 515-518.

De Datta, S.K., Bolton, F.R. and Lin, W.L., 1979. Prospects for using minimum and zero tillage in tropical lowland rice. Weed Res., 19: 9- 15.

De Datta, S.K., Lacsina, R.Q. and Ahmed, Ch.M., 1970. Water management for granular herbicides in transplanted rice. Proceedings 1st National Pest Control Conference in the Philippines, Iloilo City, pp. 231-241.

De Ploey, J. and Poesen, J., 1985. Aggregate stability, runoff generation and intenill erosion. In: K.S. Richards, R.R. Amett and S. Ellis (Editors), Geomorphology and Soils. Allen and Unwin, London, pp. 9% 120.

Diop, A.M. and Moody, K., 1984. Effect of seeding depth and flooding regime on germination and growth of Echinochloa glabrescem Philippines J. Weed Sci., 11: 65-69.

Drew, M.C., 1978. Root development and activities. In: R.A. Perry and D.W. Goodall (Editors), Arid Land Ecosystems. I. Int. Biol. Progr., Cambridge University Press, NY, pp. 573-606.

180 R.M. Bhagat et d/Agricultural Water Management 31 (1996) 165-184

Drost, D.C., 1982. Weed ecology and control in rainfed rice croppin g systems. Ph.D. Thesis, University of Wisconsin, Madison, WI, 256 pp.

Drost, D.C. and Moody, K., 1982. Effect of butachlor on Echinochlm gluhrescens in wet-seeded rice (Oryza satid. Philippines J. Weed Sci., 9: 57-44.

Eck, H.V. and Unger, P.W., 1985. Soil profile modification for increasing crop production. Adv. Soil Sci., I: 65-100.

Estominos, Jr., L.E. and Moody, K., 1983. The effect of plant spacing on weed control in transplanted rice (Oryzu s&m). Philippines J. Weed Sci., 10: 77-89.

Floresca, E.T., Calora, F.B. and Obien, S.R., 1970. Performance of pendimethalin under different water management levels in transplanted rice (Oryza scrtiuu L.). Phillip. J. Weed Sci., 6: 3 l-40.

Fryer, J.D. and Chancellor, R.J., 1970. Evidence of changing weed populations in arable land. Proceedings 10th British Weed Control Conference, Brighton, pp. 958-964.

Gajri, P.R., Arora, V.K. and Prihar, S.S., 1992. Tillage management for efficient water and nitrogen use in wheat following rice. Soil Tillage Res., 24: 167- 182.

Ghani, M.A. and Rana, S.A., 1992. Water management in paddy fields for improving irrigation system performance. Proceedings International Workshop, Soil and Water Engineering for Paddy Field Manage- ment. Asian Institute of Technology, Bangkok, pp. 330-338.

Ghildyal, B.P., 1978. Effect of compaction and puddling on soil physical properties and rice growth. In: Soils and Rice. International Rice Research Institute, Los BaZos, pp. 317.

Harwood, R.R and Bantilan, R.T., 1974. Integrated weed management. 2. Shifts in composition of the weed community on intensive cropping systems. Philippines Weed Sci. Bull., I: 37-59.

Hill, J.E., Smith, Jr., R.J. and Bayer, D.E., 1994. Rice weed control: Current technology and emerging issues in temperate rice. Aust. J. Exp. Agric., 34: 1021-1029.

Ho, N.K., Itoh, K. and Othman, A.B., 1992. Weed management and changes in rice production practices. Paper presented at the International Rice Research Conference at International Rice Research Institute, Los Banos, 21-25 April 1992, pp. 12.

Ho, N.K. and Itoh, K., 1990. The ecology of Echimchlm stcrgninu (Retz.) Beauv. and proposed control measures in the Muda area of Malaysia. Paper presented at the Third Tropical Weed Science Conference, 4-6 December 1990, Kuala Lumpur.

Ho, N.K., 1985. A brief note on Marsileu minute L. Malaysian Plant Prot. Sot. Newsletter, 9 (4): 6-7. Ho, N.K. and De Datta, S.K., 1988. Integrated weed management in irrigated rice. Proceedings 1988

International Rice Research Conference, 7- 11 November 1988. International Rice Research Institute, Los Banos, Laguna, pp. 24.

Ho, N.K. and Md. Zuki, I., 1988. Weed population changes from transplanted to direct seeded rice in Muda area of Malaysia. National Seminar and Workshop on Rice Weed 88, Penang, pp. l- 19.

Hoyle, B.J., Yamada, H. and Hoyle, T.D., 1972. Aggresizing to eliminate objectionable soil clods. California Agric., 26: 35.

Ishikawa, K., Nakamura, Y. and Kuwatsuka, S., 1975. Fate and behavior of thiobencarb herbicide in soil environment and plants. Proceedings 5th Asian-Pacific Weed Science Society Conference, Tokyo, pp. 23 l-235.

Janiya, J.D., 1984. Effect of moisture on weeds and their control in rainfed rice. M.S. Thesis, University of the Philippines, Los Banos College, Laguna.

Janiya J.D. and Moody, K., 1982. Weed control in transplanted rice (Oryza satioa) grown under different moisture regimes. Philippines J. Weed Sci., 9: 29-35.

Janiya, J.D. and Moody, K., 1983. Degree of tillage and weed control in transplanted rice (Oryzrr satiucc). Philippines J. Weed Sci., IO: 56-64.

Jikihara, K. Nakamura, Y. and Nakayama, H., 198 1. Influence of environmental factors on herbicide efficacy. Proceedings 1980 Conference on Weeds and their Control in Asia, Taipei, pp. 143-150.

Johnson, C.B., Mannering, J.V. and Moldenhauer, W.C., 1979. Influence of surface roughness and clod size and stability on soil and water losses. Soil Sci. Sot. Am. J., 43: 772-777.

Kanwar, R.S., Rao, B.N. and Murty, V.V.N., 1974. Effect of water application in paddy. Indian J. Agric. Res., 8: 145-148.

Kawamura, Y. and Hirai, K., 1975. Influence of soil properties on the herbicidal activity of oxadiazone under

R.M. Bhagut et al./Agricultural Water Management 31 (1996) 16.5-184 181

flooded conditions of paddy fields. Proceedings 5th Asian-Pacific Weed Science Conference, Tokyo, pp. 155-158.

Kim, K.U., 1981. Weeds in paddy field and their control in Korea. Proceedings 1980 Conference on Weeds and their Control in Asia, Taipei, pp. 37-50.

Kim, SC., 1980. Reduced plant spacing for weed suppression in transplanted rice. Proceedings 15th British Crop Protection Conference - Weeds, Brighton, pp. 383-387.

Kim, S.C. and Moody, K., 1980. Effect of plant spacing in the competitive ability of rice growing in association with various weed communities at different nitrogen levels. Korean Sot. Crop Sci., 25: 17-27.

Klute, A., 1982. Tillage effects on the hydraulic properties of soil: A review. In: P.W. Unger, D.M. Van Doren, Jr., F.D. Whisler and E.L. Skidmore (Editors), Predicting Tillage Effects on Soil Physical Properties and Processes. Special Publication 44, American Society of Agronomy, Madison, WI, pp. 29-23.

Kuipers, H., 1975. The objective of soil tillage. Report on the Expert Consultation Meeting on the Mechanization of Rice Production, International institute of Tropical Agriculture, Ibadan, pp. 61-65.

Lal, R. 1985a. Tillage in lowland rice-based cropping systems. In: Soil Physics and Rice. International Rice Research Institute, Los Banes, pp. 283-307.

Lal, R., 1985b. A suitability guide for different tillage systems in the tropics. Soil Tillage Res., 5: 179-196. Lo, N.P., 1988. Effects of direct seeding on weed management in Malaysia. Proceedings 2nd Tropical Weed

Science Conference, Phuket, pp. 32-46. Lubigan, R.T. and Vega, M.R., 1971. The effect on yield of competition of rice with Echinochloa crus-galli

CL.1 Beauv. and Monochoriu uaginalis (Butm f.) Presl. Philippine Agric., 55: 210-213. Mabbayad, M.O., Pablico, P.P. and Moody K., 1983. The effect of time and method of land preparation on

weed population in rice. Proceedings 9th Asian-Pacific Weed Science Society Conference, Manila, pp. 357-368.

Manuel, J.S., Mercado, B.L. and Lubigan, R.T., 1979. Approaches to the control of Paspulum distichum L. in lowland rice. Philippine Agric. 62: 255-281.

Matsunaka, S., 1983. Evolution of rice weed control practices and research: world perspective. Proceedings 1981 Conference on Weed Control in Rice. International Rice Research Institute, Los Banes, pp. 5-18.

McCaig, M., 1985. Soil properties and subsurface hydrology. In: K.S. Richards, R.R. Amett and S. Ellis (Editors), Geomorphology and Soils. Allen and Unwin, London, pp. l21- 140.

Meelu, O.P., Beri, V., Sharma, K.N., Jalota, S.K. and Sandhu, B.S., 1979. Influence of paddy and corn in different rotations on wheat yield, nutrient removal, and soil properties. Plant Soil, 5 1: 5 l-57.

Mercado, B.L., 1979. Introduction to weed science. South East Asia Regional Center for Graduate Study and Research in Agriculture, Laguna, 292 pp.

Miller, D.F. and Aarstad, J.S., 1972. Effect of deep plowing on the physical characteristics of Heizei soil. Washington Agric. Exp. Stn. Circ., 556, Pullman, WA, 9 pp.

Mism, A., Tosh, G.C. and Nanda, K.C., 198 1. Effects of herbicides and water management regimes on weeds and grain yield of transplanted rice in India. Int. Rice Res. Newsl., 6 (5): 20-21.

Moody, K., 1978. Integration of weed control methods. Regional Seminar on Integrated Pest Control for hrigated Rice in South and Southeast Asia, October 16-18 1978, Manila, 5 pp.

Moody, K., 1982. The status of weed control in rice in Asia. FAO Plant Prot. Bull., 30: 119-123. Moody, K., 1983. Weeds: Definitions, costs, characteristics, classifications and effects. In: H. Walter (Editor),

Weed Management in the Philippines. Ernst Mullerbader, Forststr 18, 7000 Stuttgart 4, pp. 11-32. Moody, K., 1985. Wet seeded rice. Women in rice farming system, International Rice Research Institute, Los

BaZos, pp. 467-480. Moody, K. and De Dana, S.K., 1980. Economics of weed control in tropical and sub tropical rice growing

regimes with emphasis on reduced tillage. Proceedings 15th British Crop Protection Conference - Weeds, Brighton, pp. 931-940.

Moturi, M., 1977. Effect of saturation and submergence during three growth stages on growth and yield of rice, direct seeded in lines on puddled soil. Thesis Abstract, S.V. Agric. College, Tirupami.

Mukhopadhyay, S.K., 1981. Weed control technology in rainfed lowland rice. In: Proceedings 1981 Confer- ence on Weed Control in Rice. International Rice Research Institute, Los Banes, pp. 109- 118.

Naphade, J.D. and Ghildyal, B.P., 1974. Influence of the state of oxidation and reduction on soil character- istics, nutrient uptake and rice growth. II Riso, 23: 277-282.

182 R.M. Bhagat et al./Agricultural Water Management 31 (1996) 165-184

Naravani, N.B. and Sirohi, B.S., 1978. Effect of tillage tools on soil structure. J. Agric. Eng., 15: 144-147. Nartsomboon, C. and Moody, K., 1988a. Effect of weeding and green manure on the yield of transplanted rice.

Proceedings 2nd Tropical Weed Science Conference, Phuket, pp. 53-77. Nartsomboon, C. and Moody, K., 1988b. Effect of seeding rate and weeding on growth and N accumulation of

Sevbania rostrata. Proceedings 2nd Tropical Weed Science Conference, Phuket, pp. 78-86. Navarez, D.C., Roa, L.L. and Moody K., 1979. Weed control in wet-seeded rice grown under different

moisture regimes. Philipp. J. Weed Sci., 6: 23-3 1. Naylor, R., 1994. Herbicide in Asian rice production: perspective from economics, ecology, and the

agricultural sciences. Paper presented at a conference on herbicide use in Asian rice production, Stanford University, CA, 28-30 March 1994.

Ng, P.H., 1986. Weed problems and control in direct seeded/broadcast rice. Proc. Weed Science in the Tropics, Unversiti Pertanian Malaysia (UPM), Serdang Selangor, pp. I 17- 120.

Noda, K., 1973. Recent changes of weed population in paddy fields. Biotrop. Bull., 11: 1979. Obermueller, A.J. and Mikkelson, D.S., 1974. Effects of water management and soil aggregation on the

growth and nutrient uptake of rice. Agron. J., 66: 627-632. Pablico, P.P. and Moody, K., 1982. Possibility of using reduced tillage for establishment of a second

transplanted rice (Oryza satioa) crop. Philippines J. Weed Sci., 9: 1 I - 17. Painuli, D.K., Woodhead, T. and Pagliai, M., 1988. Effective use of energy and water in rice-soil puddling.

Soil Tillage Res., 12: 149-161. Patcharin, W. and Prateep, K., 1988. Determination of thiobencarb and dechlorinated thiobencarb in paddy

soil. Proceedings 2nd Tropical Weed Science Conference, Phuket, pp. 296-302. Pathak, A.K., Sanlcaran, S. and De Datta, S.K., 1989. Effect of herbicide and moisture level on Rottboellia

cochinchinensis and Cyprus rotundus in upland rice. Tropical Pest Mgt., 35: 3 I l-3 15. Ponnampemma, F.N., 1972. The chemistry of submerged soils. Adv. Agron., 24: 29-96. Power, J.F., 1983. Soil management for efficient water use: Soil Fertility. In: H.M. Taylor, W. Jordan and T.R.

Sinclair (Editors), Limitations to Efficient Water Use in Crop Production. American Society of Agronomy, Madison, WI, pp. 461-470.

Prihar, S.S., Sandhu, KS., Singh, M., Verma, H.N. and Singh, R., 1989. Response of dryland wheat to small supplemental irrigation and fertilizer N in submontane Punjab. Fert. Res., 2 I : 23-28.

Prihar, S.S., Ciajri, P.R. and Arora, V.K., 1985a. Nitrogen fertilization of wheat under limited water supplies. Fert. Res., 8: l-8.

Prihar, S.S., Ghildyal, B.P., Painuli, D.K. and Sur, H.S., 1985b. Physical properties of mineral soils affecting rice based cropping system. In: Soil Physics and Rice. International Rice Research Institute, Manila, pp. 57-70.

Qu, N., 1989. Development and achievements of China’s agro-science and technology. Proceedings 1989 Asian-Pacific Weed Science Society Conference, Seoul, pp. 853-856.

Rahman, A., Mattson, B.E. and Bumey, B., 1977. Influence of selected soil properties on the phytotoxicity of soil applied herbicides. Proceedings 6th Asian-Pacific Weed Science Society Conference, Jakarta, pp. 579-586.

Raju, R.A. and Reddy, M.N., 1987. Influence of water regimes on the weed flora of transplanted lowland rice ecosystem. Andhra Agric. J., 34 (4): 459-461.

Rao, A.N. and Moody, K., 1987. Rice yield losses caused by transplanted Echinochloa glabrescens and possible control methods. Proceedings 1 lth Asian Pacific Weed Science Society Conference, Taipei, pp. 203-210.

Reddy, S.R. and Hukkeri, S.B., 1983. Effect of tillage practices on irrigation requirement, weed control and yield of lowland rice. Soil Tillage Res., 3: 147-158.

Sahu, R., 1978. Influence of soil moisture on the germination andapplication of herbicide preemergence spray to both crop and weed. Bull. Bot. Sot. Bengal, 31: 1 15-l 18.

Salokhe, V.M. and Shirin, A.K.M., 1992. Effect of puddling on soil properties: A review. Proceedings, Workshop on Soil and Water Engineering for Paddy Field Management, AIT, Bangkok, pp. 276-285.

Sanchez, P.A., 1976. Properties and management of soils in the tropics. John Wiley, NY, pp. 234-236. Sanchez, P.A., 1973. Puddling tropical rice soils. 2. Effects of water losses. Soil Sci., 115: 303-308. Sarkar, P.A. and Ghosh, A.K., 1979. Integration of weed control practices. Indian Society of Weed Science,

1979 Annual Meeting. Marathwada Agricultural University, Parbhani, p. 37 in abstract of papers.

R.M. Bhagat et al./Agricultural Water Management 31 (1996) 165-184 183

Satyal, R.P., 1978. Study on the methods of land preparation for rice cultivation. Paper presented at the Rice and Summer Crop Workshop, 27 February-2 March 1978, Department of Agriculture, Kathmandu, pp. 7.

Seth A.K., Khaw, C.H. and Fua, J.M., 1971. Minimum and zero tillage techniques and post planting weed control in rice. Proceedings 3rd Asian-Pacific Weed Science Society Conference, Kuala Lumpur, pp. 188-200.

Shad, R.A. and De Datta, SK., 1986. Reduced tillage technique for wetland rice as affected by herbicides. Soil Tillage Res., 6: 291-303.

Sharma, B.R., 1985. Response of irrigated fodder oats to nitrogen fertilization as influenced by tillage. Soil Tillage Res., 6: 69-77.

Sharma, P.K. and De Datta, S.K., 1994. Rainwater utilization efficiency in rainfed lowland rice. Adv. Agron., 52: 8% 150.

Sharma, P.K. and Bhagat, R.M., 1993. Puddling and compaction effects on water permeability of texturally different soils. J. Indian Sot. Soil Sci., 41: l-6.

Sharma, P.K. and De Datta, SK., 1986. Physical properties and processes of puddled rice soils. Adv. Soil Sci., 5: 139-178.

Sharma, P.K., Dc Datta, S.K. and Redulla, C.A., 1987. Root growth and yield response of rainfed lowland rice to planting methods. Exp. Agric., 23: 305-313.

Sharma, P.K., De Datta, S.K. and Redulla, C.A., 1988. Tillage effects on soil physical properties and wetland rice yield. Agron. J., 80: 34-39.

Sivanappan, R.K. and Saifudeen, E.S.A., 1977. Water saving method of irrigation for high yielding rice crop. Madras Agric. J., 64: 745-749.

Smith, R.J., 1983. Weeds of major economic importance in rice and yield losses due to weed competition. Weed control in rice, International Rice Research Institute, Los Banos, pp. 34-57.

Smith, Jr., R.J. and Fox, W.T., 1973. Soil water and growth of rice and weeds. Weed Sci., 21: 61-63. Soo, S.W., 1972. Semi detailed soil survey of the Kedah/Perlis coastal plain. Division of Agriculture

Ministry of Agriculture and Fisheries, Malaysia, pp. l-23. Soong, S.Y.C., 1981. New uses of molinate combinations and application techniques. In: Weed Control in

Rice, International Rice Research Institute, Los Bano. Stoskopf, N.C., 1985. Cereal gram crops. Reston Publishing, Reston, VA, 516 pp. Sur, H.S., Prihar, S.S. and Jalota, S.K., 1981. Effect of rice-wheat and maize-wheat rotation on water

transmission and wheat root development on a sandy loam of Punjab, India. Soil Tillage Res., 1: 361-371. Tabbal, D.F., Lampayan, R.M. and Bhuiyan, S.I., 1992. Water-efficient irrigation technique for rice.

Proceedings International Workshop on Soil and Water Engineering for Paddy Field Management, Asian Institute of Technology, Bangkok, 28-30 January 1992, pp. 146-149.

Tanaka, 1.. 1976. Climatic influence on photosynthesis and respiration of rice plants. Climate and rice, International Rice Research Institute, Los BaZos, Laguna.

Tauro AC., 1970. Evaluation of weed control practices in transplanted rice. M.S. Thesis, University of the Philippines, Los BaZos, Laguna.

Tisdall, J.M. and Adem, H.H., 1986. Effect of water content of soil at tillage on size distribution of aggregates and infiltration. Aust. J. Exp. Agric., 26: 193-195.

Unger, P.W., 1979. Effect of deep tillage and profile modification on soil properties, root growth, and crop yields in United States and Canada. Geoderma, 22: 275-295.

Unger, P.W., 1975. Water retention by core and sieved soil samples. Soil Sci. Sot. Am. Proc., 39: 1197-1200. Valera, A. and Wickham, T.H., 1978. A field study on water use and duration of land preparation for lowland

irrigated rice. IRRI Saturday Seminar paper, July 22 1978. International Rice Research Institute, Los BaZos, Laguna, pp. 10.

Valera, A., 1976. Field studies on water use and duration for land preparation for lowland rice. M.S. Thesis, University of Philippines, Los Banos, Laguna.

Vega, M.R. and Sierra, J.N., 1970. Population of weed seeds in a lowland rice field. Philippines Agric., 54: l-7.

Watanabe, Y. and Hirokawa, F., 1975. Ecological studies on the germination and emergence of annual weeds. 4. Seasonal changes in dormancy status of viable seeds in cultivated and uncultivated soil. Weed Res. (Japan) 19: 20-24.

184 R.M. Bhagut et al./Agricultural Water Manqemrnt 31 (1996) 165-184

Williams, J.F., Roberts, S.R., Hill, J.E., Scardaci, S.C. and Tibbits, G. 1990. Managing water for weed control in rice. California Agric., 441 7-10.

Wrigley, G., 1969. The problem of weeds in rice. Rice Tech. Monogr. No. 1, CIBA, Basle, pp. 27-3 I. Zimdahl, R.L., Moody, K., Lubigan, R.T. and Castin, E.M., 1988. Patterns of weed emergence in tropical

soils. Weed Sci., 36: 603-608. Zimdahl, R.L. Moody, K. and Chavez, R.C., 1987. The influence of soil moisture on growth of some rice

(Oryzo sutiua) weeds. Philippines J. Weed Sci., 14: 19-25.