Search Research Database
Project Record

Objectives:

To develop and promote improved and cost-effective methods for the control of weeds in flood plain rice production systems, through an understanding of the factors that determine variability in weed communities and their persistence, to provide a basis for the integration of fallow management and in-crop weed control practices which impact on the long term persistence of weeds in the weed seed bank.

Background:

Rice production is crucial to the economy of Bangladesh. Approximately 75% of the cropped area is devoted to rice production, with some 60-70% of the agricultural labour employed in rice production, marketing and distribution. The present rate of population increase is almost 2% per annum, while the area available for rice production is declining, so that large gains in productivity will be needed in order to sustain domestic production. Improved technologies are needed to increase returns for resource-poor farmers, while at the same time protecting the resource base for future generations. Weeds are cited as among the most severe biotic constraints, and losses attributed to pests and diseases are seen to be less severe in rainfed systems than in irrigated rice production systems. Rainfed rice systems are dominated by the need for the timely establishment of the crop with the onset of rains, which typically shortens the time available for good land preparation and weed control. Traditional rice varieties predominate in rainfed lowland systems, photoperiod sensitivity in most varieties determines that flowering occurs at a specific time, usually assuring that the crop avoids entering the reproductive stage during a period of high risk of drought. However, the very specific photoperiod requirements for each locality limits the value of germplasm exchange, and breeding for weed suppressive crop cultivars has been delayed. Although they usually give some yield over the range of environments encountered, traditional varieties often respond poorly to more intensive management and inputs. When fertilisers are applied, they usually respond with excessive vegetative growth and often lodge near harvest time. Immediate gains in crop protection through weed management are therefore likely to be achieved through research into appropriate agronomic interventions, in addition to the selection of weed suppressive cultivars. The project outlined in this memorandum will characterise the variability in weed communities in rainfed lowland rice cropping systems in Bangladesh, examine the persistence of weed seeds in the soil reservoir, and investigate in on-farm trials the role of fallow management and stale seed bed techniques for managing weed populations.

Intended Outputs:

Crop and weed management practices and farmers' perceptions of weeds and control measures described.

Weed populations within different ecologies and the influence of crop and weed management practices decribed.

Persistence, dormancy and quantity of weeds seeds in the seed bank determined.

Integrated weed management strategies to reduce the build-up of weed populations, by regulation of soil seed banks which are appropriate to the farming system and cultural context of each area studied, developed.

Findings disseminated through IRRI Weed Ecology Working Group, and at end of the project via a workshop for NGOs and extension staff.

Progress and Impact:

Crop and weed management practices and farmers perceptions of weeds and control measures:
Following an inception meeting of project partners two areas were selected in which to conduct field studies. These provided the contrast of a predominantly single monsoon (T-aman) rice crop/fallow system in the Barind Tract of Rajshahi district with more intensive multiple rice cropping systems of T-aman/boro (monsoon/irrigated rice crop) or T-aman/boro/T.aus (monsoon/irrigated rice/pre-monsoon rainfed rice crops) in Comilla district. A methodological mix of structured questionnaire surveys, focus group discussions, and household case-studies allowed an in depth analysis of farmers' weed management. Labour, farm size, land tenure, and district (reflecting knowledge of new rice technology) were shown to be important determinants of the timeliness of weeding. To facilitate the design of appropriate interventions it was necessary to understand a range of components of farmer weed management decision making. These include the effects of land tenure on weed management, the importance of water control for weed management in boro rice, the likely impact of herbicide adoption on rural employment, the use of weeds to feed livestock, the scope for dryland (rabi) cropping in the off-season in the Barind and, the potential profitability of new weed management practices.

Weed populations within different ecologies and the influence of crop and weed management practices:
A study of yield gaps due to weeds, under farmer weed management practices, has demonstrated that many farmers fail to realise substantial proportions of the rice yield due to weed competition. Further more sub-optimum weed control can also prevent the full benefit from fertiliser inputs being realised. In the T-aman approx. 30% of farmers could save losses of 0.5 t ha-1 or more by investing in additional weeding. In Comilla the yield gap due to weeds in boro rice averaged 0.45 t ha-1. In the aus, when there was no significant difference in yields from unweeded plots and those managed by farmers, additional weeding resulted in a mean yield increase of 1.1 t ha-1. Monitoring of farmer managed crops provided a listing of the dominant weeds at different positions on the toposequence. In Rajshahi fields are infested by three potentially difficult to control rhizomatous grasses, Cynodon dactylon, Paspalum distichum and Leersia hexandra. Based on experience elsewhere, these perennials, a number of other grasses including Echinochloa spp. and Ischaemum rugosum, and sedges, particularly Cyperus difformis will pose a threat to yields in a direct seeding system too dependant on herbicide use.

Persistence, dormancy and quantity of weed seeds in seed bank determined:
A study of the soil seed bank, under a range of weed management practices in a long-term trial at Rajshahi, was used to determine the effect that a change in rice establishment, from transplanting to direct seeding, may have on weed abundance and species composition. This work has identified the diversity of the weed flora present in the soil and indicated the periodicity of germination which may be expected and the speed of response in relation to flooding.

Integrated weed management strategies to reduce the build-up of weed populations by regulation of soil seed banks which are appropriate to the farming system and cultural context of each area studied developed:
Fully replicated trials were undertaken at the Rajabari (Rajshahi) key site of Rainfed Lowland Rice Research Consortium in aman 2000 and 2001 to compare rice yield and weed population response to different rice crop establishment, nutrient management and weeding practices. The effects of timing of pre-plant tillage and in-crop weed control by herbicides, mechanical and hand weeding were evaluated for direct-seeded rice. Clear system effects and interactions between cropping and weed management practice have been observed in a

Project Conclusions:

The project has demonstrated that improved weed control needs to be an essential component of pest management if average yields of lowland rain-fed rice are to be increased. By building upon earlier work in Bangladesh and through a study of the effectiveness of weed management within existing systems, the project has contributed to an understanding of what the elements of cost-effective weed management strategies need to be. Project findings can now be used as the basis for promoting improved weed management methods in the single rice aman crop and intensified multi-crop aman-boro and aus-aman-boro systems. Successful weed management results when farmers are able to integrate a set of crop management decisions which lead to the establishment of a vigorous, competitive crop in which timely in-crop weed control is achieved. Knowledge on the effects weeds on the crop, timeliness of operations, how to use individual control practices effectively and the long term impact of cropping practices on weed populations themselves is key to effective farmer decision making. The outputs contribute to the project goal by providing information on these issues in a form that will be accessible to extension and NGO workers to use in farmer training. The project has identified that increasing use of herbicides in intensive rice systems in Bangladesh is inevitable. As herbicides are inexpensive, and reduce input costs over hand weeding, widespread adoption by resource poor households is also likely. With adequate training the herbicides that are now on the market in Bangladesh can be used safely. Farmer access to knowledge is the key to ensuring that herbicides are used as a component of an integrated, environmentally benign pest control programme in lowlands of Bangladesh. The working paper on herbicide adoption circulated to and discussed with the agrochemical companies that supply the Bangladesh market has started the process of raising awareness of the issues on which farmers need additional information.

General Notes:

See also the TECA Record Weed management practices in intensive irrigated rice and rain-fed rabi systems.

Publications:

Riches CR (1999). Visit to Bangladesh to establish project. 16-21 October 1999. Back to office report. Natural Resources Institute (NRI), University of Greenwich, UK. 3 pp (C)



AHMED, G.J.U., HASSAN, M.A., MRIDHA, A.J., JABBAR, M.A., RICHES, C.R., ROBINSON, E.J.Z. and MORTIMER, M. (2001) Weed management in intensified lowland rice in Bangladesh. pp. 205-210 In: Proceedings of the Brighton Crop Protection Conference, Weeds 2001. Farnham, UK: British Crop Protection Council.



MAZID, M.A., JABBAR, M.A., RICHES, C.R., ROBINSON, E.J.Z., MORTIMER, M. and WADE, L.J. (2001) Weed management implications of introducing dry-seeding of rice in the Barind Tract of Bangladesh. pp. 211-216 In: Proceedings of the Brighton Crop Protection Conference, Weeds 2001. Farnham, UK: British Crop Protection Council.



MAZID, M.A., WADE, L.J., MORTIMER M. and RICHES C.R. (Eds.) (2001) Improving lowland rice production and weed management. Proceedings of the annual review and planning workshop held 27th to 28th March 2001, Bangladesh Rice Research Institute, Gazipur, Bangladesh and International Rice Research Institute, Philippines.



MAZID, M.A. (2001) Presentation on the project by M. A. Mazid in the International Rice Research Institute Rainfed/Upland Rice Research Consortium 12th Steering Committee Meeting, 10-11 December 2001 at Central Rice Research Institute, Cuttack, India.



JABBAR, M.A., AHMED, G.J.U., MAZID, M.A., ORR, A.W., ROBINSON, E.J.Z. and RICHES, C.R. (2002) Weed management in floodplain rice in Comilla and Rajshahi Districts, Bangladesh: Characterisation Studies. Working Paper. Natural Resources Institute, University of Greenwich, Chatham, UK. 44 pp.



ANON (2002) One lakh additional tons of rice can be produced in Barind land. Newspaper article reporting work undertaken at Bangladesh Rice Research Institute. The Bangladesh Observer, 14 February 2002



DANQUAH, E. Y., JOHNSON, D.E., RICHES, C., ARNOLD, G.M. and KARP, A. (2002) Genetic diversity in Echinochloa spp. collected from different geographic origins and within rice fields in Côte d'lvoire. European Weed Research Society Weed Research 2002, 42: 394-405



RICHES, C.R., AHMED, G.J.U., BADSHAH, M.A. and BHUIYAN, M.K.A. (2002) Herbicide adoption in Comilla District, Bangladesh. Working paper for DFID Crop Protection Programme project R7471. Natural Resources Institute (NRI), Chatham, UK. 27pp.



ORR, A.W. and JABBER, M.A. (2002) Farmers' weed management for T.aman, Bangladesh. Working paper for DFID Crop Protection Programme project R7471. Natural Resources Institute (NRI), Chatham, UK. 27 pp.



ORR, A.W. and JABBER, M.A. (in press) Expanding rabi cropping in the High Barind Tract, Bangladesh: a socio-economic perspective. Working paper for DFID Crop Protection Programme project R7471. Natural Resources Institute (NRI), Chatham, UK. 27 pp.



Uddin, J. 2002. Research issues in weed management in Bangladesh. Paper presented at the Annual Steering Committee Meeting, Irrigated Rice Research Consortium, Bangkok January, 2002. International Rice Research Institute (In press).



Mazid M. 2002. Dry seeding rice for enhanced productivity of rain-fed drought prone lands: lessons from Bangladesh and the Philippines. In : Pandey, S., Mortimer; M., Wade, L., Tuong, TP and Hardy, B (2002). ,Direct seeding in Asian rice systems: strategic research issues and opportunities,. International Rice Research Institute (In press 383 pp)

Associated References:

R6751, R7296