Previous Page Table of Contents Next Page

The need for improved weed management in rice - R. Labrada

Weed Officer, Plant Protection Service, FAO, Rome, Italy


Weeds are the cause of serious yield reduction problems in rice production worldwide. Losses caused by weeds vary from one country to another, depending on the predominant weed flora and on the control methods practised by farmers. Two examples give an idea of the dimensions of the problem. In China, 10 million tonnes (Mt) of rice are lost annually due to weed competition (Ze Pu Zhang, 2001); such a quantity of rice is sufficient to feed at least 56 million people for 1 year. In Sri Lanka, a country considered self-sufficient in rice, weeds are the major biotic stress in rice production and account for 30 to 40 percent of yield losses (Abeysekera, 2001).

Weed control in overpopulated areas of Asia has mainly been carried out through a combination of water management and hand-weeding, but the latter is becoming less common in areas with an increasing labour shortage problem; furthermore, this method affects transplanting. For these reasons, many farmers in several regions of the world, including Asia, have shifted from transplanting to direct seeding rice; less labour is required but herbicides must be used for weed control. Farmers are then faced with no option other than the application of herbicides, despite their lack of knowledge concerning the proper use of these chemicals.

Herbicide-based weed management is becoming the most popular method of weed control in rice. However, while herbicide application certainly controls several weeds, it does not eliminate others, thereby provoking a weed shift of tolerant species. In some areas it is believed that herbicide use will solve all weed problems. Experience shows, however, that although herbicide use alleviates the problem of labour for weeding, incorrect use of herbicides may bring about other environmental problems. The advent of herbicide-resistant species is an increasingly worrying problem for farmers, extension workers and policy-makers in many rice-producing areas in Asia and Latin America.

The only way to avoid these problems is the implementation of improved weed control within the context of integrated pest management, with particular emphasis on the weed ecobiology of the prevailing species. This is an important prerequisite for achieving the expected yield growth in rice production and obtaining the necessary reduction in weed stand, including weed seed bank.

This paper aims to give an idea of the work entailed in the implementation of improved weed management practices.


The main weed control method in rice in Asian countries has until recently been manual weeding combined with water management. Nearly all countries in the region used transplanting as a planting method. In Latin and North America, chemical control has been used for 30 years for both transplanted and direct-seeded rice; the main herbicide treatment has been propanil applied post-emergence.

Labour shortage is an increasing problem in several Asian countries. Even overpopulated countries, such as China, face this problem; for other nations, such as the Republic of Korea and Japan, where mechanical transplanting is common practice, the problem may be felt in years to come. This trend points towards an increase in direct-seeded areas, where rice plants are less competitive with weeds. In some countries (e.g. Malaysia and Viet Nam), direct seeding is already practised and the use of herbicides is unavoidable.

In the western hemisphere, many farmers normally use herbicides in rice cultivation, but experience over the past 30 years has in turn created problems:

Experience shows that although herbicide use has increased productivity, there are several weed problems that remain unsolved and for which other solutions need to be developed and implemented (Labrada, 1996).

Work over the last 10 years has identified several weeds affecting rice production, and in particular two major weed complexes requiring special attention and an integrated approach for their control: the complex formed by Echinochloa species and red/weedy rice.

Echinochloa weed species are a major constraint to rice production worldwide. These weeds are widely associated with rice and the damage they cause to the crop is permanent. Echinochloa crus-galli is found in many countries, except Central America; E. colona is present almost everywhere in hot climate countries; and other species exist which are confined to particular environments. Echinochloa spp. have the capacity to adopt mimetic forms to rice, with the result that their control by manual weeding is very difficult. Red/weedy rice is a problem brought about by the increase in area of direct-seeded rice. Both weed problems are the result of permanent monocropping in rice.

Herbicide resistance is a major problem, particularly the abundance of propanil-resistant Echinochloa colona in Central and South America and E. crus-galli populations in the United States of America. Other weed species, such as Eleusine indica, have also developed resistance to the inhibitors of acetyl CoA carboxylase (ACCasas) in Malaysia, while Monochoria vaginalis has evolved resistance to ALS inhibitors (Heap, 2001). Other weed species may also develop resistance in the future making the improvement of weed control more difficult.

Table 1 lists major weed problems by country; it is clear that some weed problems are common to more than one country while others are more specific, depending on the agrotechnical management of the crop. For example, rainfed rice is normally affected by species such as Rottboellia cochinchinensis, Cynodon dactylon and Cyperus rotundus, while other species prevail in flooding conditions.

Weed species reported as important in rice in various countries of the world



Echinochloa crusgalli

Aus, BG, Ch, Egy, Indo, Iran, Jap, Kor, Nep, Mal, SL, Sp, Th, USA, Vie

E. colona

Aus, Bel, BG, Co, Con-D, CR, Cu, Gh, Gy, Indo, Mal, Mex, Nep, Nic, Pan, Per, SL, Th., Ven, Vie

E. glabrescens


E. oryzicola

Jap, Kor

Eleusine indica

Bel, Con-D, Mal

Ischaemum rugosum

CR, Cu, Mal, Nep, Ven

Leptochloa chinensis

Ch, Th, Vie

Rottboellia cochinchinensis

Bel, Col, Con-D, Pan, Per

Cynodon dactylon


Scirpus planiculmis


Sagittaria pygmaea Miq


Potamogeton distinctus


Paspalum distichum

Ch, Con-D, Iran, Nep, Pak

Ludwigia prostrata


Monochoria vaginalis

Ch., Iran, Kor, Vie

Scirpus juncoides


Cyperus difformis

Ch, CR, Cu, Egy, Gh, Gy, Nep, Pak, Th, Vie

Cyperus iria

CR, Cu, Egy, Gh, Gy, Indo, Iran, Nep, Pak, Th, Vie

Cyperus rotundus

Bel, BG, Con-D, Egy, Mex, Pak

Eclipta prostrata L.

Ch, CR, Cu, Egy, Indo, Mex, Nep, Th, Vie

Fimbrystilis littoralis

Gy, Nep, Pak

Red/Weedy rice

Bel, Bra, Col, CR, Cu, Fr, Gy, It, Mal, Mex, Nic, Por, Sur, USA, Ven, Vie

a Aus - Australia; Bel - Belize; BG - Bangladesh; Bra - Brazil; Ch - China; Col - Colombia; Con-D - Democratic Republic of the Congo; CR - Costa Rica; Cu - Cuba; Egy - Egypt; Fr -France; Gh - Ghana; Gy - Guyana; Indo - Indonesia; It - Italy; Jap - Japan; Kor - Republic of Korea; Mal - Malaysia; Mex -Mexico; Nep - Nepal; Pak - Pakistan; Pan - Panama; Per -Peru; Por - Portugal; SL- Sri Lanka; Sp - Spain; Sur -Suriname; Th - Thailand; USA- United States of America; Ven - Venezuela; Vie - Viet Nam.

The traditional approach to weed control has been to rely on one single main method which is then combined with other procedures. In the past, the main method was land preparation plus manual weeding, including some fallow. With the advent of herbicides, farmers in many areas began replacing manual weeding with chemical control while fallow diminished, mainly due to land pressure and demographic growth, particularly in Asia. In order to improve weed management, appropriate measures must be taken on the basis of specific studies of:

Other tools for weed control also need to be identified:


Knowledge of the behaviour of weed species is lacking in the basic information available for weed control in most developing countries. Nevertheless, the interaction of weeds as a complex and their seed bank is an important aspect requiring attention. Control measures have always been adopted to reduce weed infestation at certain phases of the crop cycle, but never in order to bring about a sustainable reduction of the infestation.

In order to know when is the right time to implement control methods to reduce weed species productivity, knowledge is required concerning: weed productivity; time of germination/emergence; and the period of fruit-setting and/or emission of first vegetative organs. These studies also provide information on the negative or positive influence of certain biotic and abiotic factors on weed growth and development (Figures 1 and 2). Weed management must focus on the control of weed species, not only to avoid competition but also to prevent further build-up of weed seed bank in soil and eventually reduce it.

Components of Integrated Weed Management bio-ecology of weeds (population dynamics, seed bank)

Cycle of a single annual weed species

It is true that entomologists and pathologists have to work with a small number of pest species, while weed scientists deal with a complex that may easily have more than 10 species in a single site. However, tables of weed life, including most appropriate conditions for germination and diaspora production (at least of the most important species), are a suitable guide for farmers to improve weed control. Farmers should also pursue not merely the control of weeds but their reduction every year. Simple methods to evaluate viable weed seed bank can be implemented at farmer level.


There are already 252 resistant weed biotypes belonging to 154 species (of which 92 are dicots and 62 monocots) infesting more than 200 000 fields (International Survey of Herbicide Resistant Weeds, In the case of rice, besides the problems of Echinochloa spp. with propanil, other resistant species have been detected (Table 2).

Weed species resistant to herbicides

Herbicide group or herbicide



Inhibitors of lipid synthesis
(butachlor, thiobencarb, molinate)

Echinochloa crus-galli

China, USA, Thailand

Phenoxy herbicides (2,4 D)

Fimbrystilis miliacea, Sphenoclea, zeylanica, Limnocharis flava

Malaysia, Philippines

Inhibitors of ACCase
(Arylophenoxy propionates)

Echinochloa colona

Costa Rica, Colombia, Nicaragua

Inhibitors of ACCase
(Arylophenoxy propionates)

Ischaemum rugosum



Echinochloa crus-galli

Spain, USA

Echinochloa colona


Echinochloa crus-pavonis


Sulphonyl ureas

Monochoria korsakowii

South Korea

Sulphonyl ureas

Rotala indica


On the basis of experience in various countries, the approximate timing of and intervals between applications which result in problems of resistance are known for each herbicide group or single compound. It is, therefore, only natural to propose the establishment of guidelines for the use of the herbicides to prevent resistance.

Valverde and Itoh (2001) propose two strategies:

In rice there is an urgent need to implement control measures because several fields are sown twice or even three times a year, with the result that they may receive three or more applications of the same herbicide in one year. Selection pressure of the herbicide therefore increases every year, shortening the period for resistance to develop and thrive in the field. It is important for farmers to know when resistance may appear and the likelihood of the species becoming resistant; they are then in a position to know when the acquired resistance will disappear. Research in this area is essential: a resistant species may become as harmful as a pest in the crop, and farmers need appropriate advice. In the pesticide registration scheme, at the time of approval of any new herbicide for use, it is important to request available information on this matter from the supplier so that it can be used later by extension workers in their farmer training.

Farmers should also be trained in the correct use of herbicides and provided with the necessary know-how to prevent resistance. Labelling of herbicides should include the correct mode of action - information which can help reduce the possibilities of herbicide resistance.


Progress has been made recently in weed management. Results have tended not to consider the ecology of the weeds concerned; nevertheless, some of the procedures developed may help to bring about more environmentally safe weed control in rice.

Crop rotation with allelopathic crops and rice cultivars

Some crops, such as sorghum, pearl millet and maize, may drastically suppress the weed population and reduce its biomass. Pearl millet may exhibit residual weed suppression in the following crop. The inclusion of these fodder crops before the rice crop in a rice-wheat rotation may provide satisfactory weed control and can minimize the use of herbicides (Narwal, 2000). It is obviously necessary to evaluate whether these crops can be grown successfully in lowland areas.

Rice allelopathy has been on the IRRI research agenda for a decade. Results have shown that there is wide variation in allelopathy among rice cultivars. Allelopathic rice can suppress both monocot and dicot weed species (Olofsdotter, 2001). Planting of barnyard grass (Echinochloa crus-galli) together with various varieties of rice in the greenhouse has shown the potential of some varieties to inhibit weed growth by up to 40 percent (Mattice et al., 1999). Similar research is taking place in the Republic of Korea (Lin et al., 2000; Cho-Young Son et al., 1999), where there is potential for improved weed management in the near future.

Competitive rice cultivars

Studies have examined the competitive ability of rice cultivars with the weeds Echinochloa colona, E. phyllopogon, E. oryzoides, Brachiaria brizantha or B. decumbens (Fischer and Gibson, 2001). The results indicate that it is not necessary to develop high erect cultivars (normally susceptible to lodging) in order to achieve a high level of competitiveness; modern high-yielding, semi-dwarf cultivars are also able to compete with weeds efficiently.

Competitive rice cultivars (e.g. semi-dwarf cv. M-2002) effectively suppressed the infestation of Echinochloa oryzoides and E. phyllopogon in California and may help reduce herbicide dependency and decrease selective pressure for resistance (Gibson et al., 2001).

Cycle of a perennial weed species

According to Fischer and Gibson (2001), competition for light is a critical factor in the process of interference between rice and weeds. Leaf area and number of tills are characteristics directly correlated with the capacity of the crop to intercept light and suppress weed growth. This suggests the importance of combining phenological characteristics to maximize the level of competitiveness of rice with weeds. A strong crop root system also enables the plant to compete at early stages with weeds when the leaf of the crop is yet to be developed.

Weed-smothering with green manure

Rice may be successfully intercropped with legumes, such as Crotalaria juncea, Vigna sinensis, Glycine max and Sesbania rostrata. Experiments conducted in India show that rainfed lowland rice intercropped with any of the above-mentioned crops’ yields as much as rice treated with the herbicide butachlor (Angady and Umapathy, 1997). Intercropping also smothers weed stand composed of Echinochloa colona, Panicum spp., Ischaemum rugosum, Cyanotis spp. and Eclipta prostrata.

Chinese milk vetch (Astragalus sinicus) can also be used in rice-vetch relay cropping systems (Cho-Young Son et al., 1999) in no-till direct-seeded rice. This method allows weed stand to be markedly reduced thanks to the shading provided by the vetch. The advantage of this method is that vetch degrades quickly with flooding

Rice seeding

While increasing rice seeding rates increases weed control, studies reveal that increased seed rates may bring about problems of disease and mutual shading. Furthermore, some water weeds (e.g. Monochoria vaginalis, Sphenoclea zeylanica, Utricularia aurea and Nymphoides indica) are not affected by high seed rates of 200-300 kg/ha (Rothuis et al., 1999). The use of high seed density should be reconsidered within the context of Integrated Crop Management to evaluate its economical feasibility and compatibility with other aspects of cropping.

Control of troublesome weed complexes

Echinochloa spp.

Information from the workshop held in China (FAO, 2001) indicates that preventative measures should be used, particularly field hygiene, to arrest the spread of Echinochloa seeds and propagules from the source of infestation to unaffected areas. Tillage, water management and the use of selective herbicides are the best combination for field control of Echinochloa spp. Over-reliance on a single herbicide may lead to herbicide-resistant biotypes; monitoring of herbicide-resistant Echinochloa populations is therefore considered essential for improved weed management in areas of rice monocropping.

Red rice

In Brazil, a new rice production system has been introduced, consisting of two early irrigation sessions during the rice cycle in order to promote red rice weed germination. Immediately after germination of weeds, glyphosate is applied in two sprayings: first, alone at a low rate of 0.384 kg a.i./ha (or mixed with 2,4 D [0.36 kg a.i./ha] in the presence of broadleaf weeds); and second, at a reduced glyphosate rate of 0.240-0.288 kg a.i./ha. This new system proved equal or superior to the traditional system of red rice control and reduced the soil seed bank considerably (Foloni, 1999).

The global workshop on red/weedy rice control (FAO, 1999) recommended the application of an integrated approach. Akey control measure is the use of clean rice seeds and any certified rice seeds should be free of weedy/red rice seeds. In the field, wet soil preparation (to bring about the emergence of weedy/red rice seeds) should be combined with the application of herbicide (e.g. glyphosate) over the weed stand and before rice seeding, plus water management before and after seeding. Minimum or zero tillage combined with the use of pre-planting non-selective herbicides is recommended in upland and irrigation conditions.

From the above it is clear that crop rotation, proper tillage, water management and use of clean seeds and machinery, together with rational herbicide use, are the main components of integrated weed management in rice.


Some transgenic crops resistant to broad-spectrum herbicides have been developed and their commercial release is imminent, in particular for imidazolinone- and glufosinate-resistant varieties in the United States and Latin America (Olofsdotter et al., 2000). Rice is amongst these transgenic crops. The idea is to use a post-emergence herbicide of broad-spectrum action in order to eliminate troublesome weeds at an early stage of the crop. This may improve weed management and crop yields, but there is concern about the environmental impact of herbicide-resistant rice, in particular about the possibility of transfer of the resistance trait to red/weedy rice. Such a situation could prove more problematic than the current one in direct-seeded areas. At present, however, farmers are simply unable to afford such approaches. The idea that resistance may evolve in 4 to 5 years and that in the same period other herbicide-resistant rice may be developed is both simplistic and risky. There is no doubt that science may do something to prevent this problem in the future, but at present no solutions are available.


The Global IPM Facility (2002) has stated that an integrated pest management programme should begin with the question: “Given the constraints under which a farm community is working, what can be done to improve pest management?” The appropriate response is to obtain ecological information on crops and weeds and their interaction (competition and allelopathy). Based on this information, a control approach should be developed in order to reduce weed infestation not only at the time of the critical periods of competition but also at a time that will prevent further build-up of weed seed bank in soil. The approach should try to minimize weed infestation and herbicide dependency. It should also have elements for preventing herbicide resistance. Therefore, it is imperative that weed management elements be urgently included in the IPM curriculum. Incorporation of such knowledge will facilitate farmers’ training in ongoing IPM projects in rice in Asia. In addition, research should be able to study further important issues, such as:

This information on weed ecology should be provided regularly to extension workers who will subsequently work in farmers’ training on improved weed management.



Abeysekera & Anuruddhika, S.K. 2001. Management of Echinochloa spp. in rice in Sri Lanka. Paper presented at the FAO workshop on Echinochloa spp. control, Beijing, China, 13 pp.

Angady, V.V. & Umapathy, P.N. 1997. Integrated weed management through smother intercrops in rainfed lowland rice. International Rice Notes, 22(1): 47-48.

Cho-YoungSon & Choe-ZhinRyong. 1999. Vetch effects for the low-input no-till direct-seeding rice-vetch cropping system. Korean J. of Crop Sci., 44(3): 221-224.

FAO. 1999. Global workshop on red rice control. Report, Varadero, Cuba, 30 Aug. - 3 Sept., 155 pp.

FAO. 2001. Summary of the Workshop on Echinochloa Control, Institute of Plant Protection, Beijing, China, 27 May. Available at

FAO. 2002. Global IPM Facility, Doing better science together. Available at

Fischer, A.J. & Gibson, K.D. 2001. Cultivares competitivos como herramienta para el Manejo Integrado de Malezas. Resúmenes XV Congreso ALAM, Maracaibo, Venezuela, 26-30 Nov., p. 71-72.

Foloni, L.L. 1999. New weed management system in no till irrigated rice aiming to improve red rice control. Proceedings of Brighton Crop Protection Conference: Weeds, Brighton, UK, 15-18 Nov. 1999. Vol. 1, 369-372.

Gibson, K.D., Hill, J.E., Foin, T.C., Caton, B.P. & Fischer, A.J. 2001. Water-seeded rice cultivars differ in ability to interfere with watergrass. Agron. J., 93(2): 326-332. Heap, I. 2001. International survey of herbicide resistant weeds. Available at

Labrada, R. 1996. Weed management in rice. In Auld, B.A. & Kim, K.U. eds. FAO Plant Production & Protection Paper No. 139, p. 259-272, FAO, Rome.

Lin, W.X., Kim, K.U. & Shin, D.H. 2000. Rice allelopathic potential and its modes of action on barnyardgrass (Echinochloa crus-galli). Allelopathy J., 7(2): 215-224.

Mattice, J.D., Skulman, B.W., Dilday, R.H., Moldehauer, K. & Lavy, T.L. 1999. Project report on chemical aspects of rice allelochemicals for weed control, Feb. 1999. Research Series, Arkansas Agricultural Experiment Station, No. 468, 90-96.

Narwal, S.S. 2000. Weed management in rice: wheat rotation by allelopathy. Critical Reviews in Plant Sci., 19(3): 249-266.

Olofsdotter, M. 2001. Rice - a step toward to use allelopathy. Agron. J., 93(1): 3-8.

Olofsdotter, M., Valverde, B. & Madsen, K.H. 2000. Herbicide resistant rice (Oryza sativa L.): global implications for weedy rice and weed management. Annals of Applied Biology, 137(3): 279-295.

Rothuis, A.J., Vromant, N., Xuan, V.T., Richter, C.J.J. & Ollevier, F. 1999. The effect of rice seeding rate on rice and fish production, and weed abundance in direct-seeded rice-fish culture. Aquaculture, 172(3-4): 255-274.

Valverde, B.E. & Itoh, K. 2001. World rice and herbicide resistance. In Powles, S.P. & Shaner, D.L. eds. Herbicide resistance and world grains, p. 195-249, New York, CRC Press.

Ze Pu Zhang. 2001. Weed management in rice in China. Summary presented at FAO Workshop on Echinochloa spp. Control, Beijing, China, 27 May.

Previous Page Top of Page Next Page