Input-responsive, high yielding rice varieties and associated technologies responsible for the doubling of yields on irrigated lands in Asia have not suited the area of more than 40 million hectares of Asian rainfed lowland rice. These environments are home to some of the poorest rural populations in South and Southeast Asia, and the rice crops are subject to drought, prolonged submergence from uncontrolled flooding and nutrient deficiencies. Farmers grow unimproved varieties and these, combined with abiotic stresses and low inputs, result in grain yields often less than 2 t ha−1. The International Rice Research Institute (IRRI) and National Agricultural Research Systems (NARS) have recently joined to form the Rainfed Lowland Rice Research Consortium in order to identify, prioritize and execute strategic research that addresses critical yield and productivity constraints. Principal intervention points for achieving sustainable yield increases are in developing drought- and submergence-tolerant germplasm with good yield potential, improved nutrient management under stress conditions, water use-efficient crop establishment practices, and understanding farmers' approaches to risk management. Multidisciplinary teams of IRRI and NARS scientists execute research at sites selected across the region to represent the key sets of constraints.

Land and crop management practices were studied to determine yield stability and feasible measures to improve productivity of dry-seeded rice. Variability of yields among farms that had grown dry-seeded rice on the entire farm every season during the period 1991–95 was lower than among farms that had consistently grown transplanted rice during the same period. Weed control is a major challenge for dry-seeded rice and almost all farmers used herbicides to control weeds. Farmers who grew a non-rice crop before the rice season had better weed control than those who did not. Ploughing intensity and the type of implement used for land preparation were not significant factors for weed control. Field bund management was one determinant of productivity differences among farmers. The use of cross-bunds or periphery- and cross-bunds improved the efficiency of rainwater conservation and input use, and increased yields. Fields with higher levelling precision had lower water stress and produced better yields. On average, a farmer lost 0.93 t ha−1 yield due to land-levelling deficiency. The division of large and medium farms into a number of smaller plots improved levelling precision and water control, resulting in higher yields.

Floodwater was characterized through field surveys over three years in rainfed lowland and deepwater rice areas of Eastern India. Measurements focused on dissolved O2 and CO2, pH and irradiance during flash floods in rice fields. Over locations and time, dissolved O2 concentrations ranged from zero to 0.28 mol m−3 (0–1.1 times air-saturated water at 30 °C) while dissolved CO2 ranged from 0.28 to 1.96 mol m−3 (31–217 times air-saturated water). Floodwater pH varied from 6.6 to 9.7. Irradiance decreased with depth in the water profile to an extent depending on turbidity. Turbidity varied greatly over locations and time. The significance of these measurements in assessing rice submergence tolerance is discussed.

A study was conducted from October 1994 to March 1996 to assess groundwater dynamics and quality in relation to landuse and farm input of nitrogen fertilizer in a highly diversified and intensive agricultural area at Magnuang, Batac, Ilocos Norte. Monthly groundwater depths, nitrate-nitrogen (NO3-N), chloride (Cl−), bicarbonate (HCO3−), electrical conductivity (EC) and pH were determined in 19 agricultural and domestic wells. In the dry season, sweet pepper (Capsicum annuum) had a higher irrigation requirement and caused more groundwater level decline than other crops. EC (700–3000 μmho cm−1) and HCO3− (90–500 ppm) in all wells exceeded the FAO threshold quality for irrigation but were not related to farm management practices. Eight wells showed near or above the World Health Organization NO3-N limit (10 ppm) for drinking water. High nitrogen fertilizer input increased the mean NO3-N (r2 = 0.45, p < 0.002). The percentage of the wells' service area under rice cultivation in the wet season accounted for 84% (p < 0.001) of the variation in NO3-N among the wells. The mean NO3-N declined as the percentage of service area under rice increased. This was related to the denitrification process in the flooded fields and the lower levels of nitrogen fertilizer for rice compared with other crops.

Low concentrations of phosphorous were added to the floodwater of submerged rice in an attempt to regulate algal growth, so as to increase floodwater O2 concentration and plant survival during submergence for up to 12 days. Increasing the phosphorus concentration of the floodwater by 1 ppm increased algal growth by at least 4-fold based on chlorophyll concentrations, and increased floodwater O2 concentrations to over twice air saturation. However, additions of phosphorus reduced rice plant survival during submergence by up to 35%. The adverse effects of a high phosphorus concentration in the floodwater were mainly attributed to competition between the algae and the submerged rice for CO2 and, perhaps, light. The importance of photosynthesis during the submergence of rice was supported by the results of experiments in which floodwater CO2 concentration was manipulated by altering pH. The survival of an intolerant rice cultivar during submergence increased from 0 to 17 and 62% at pH 8, 7 and 5 respectively, while floodwater CO2 concentrations at these pHs would have increased from 0.02 to 0.3 and 1.0 mol m−3 respectively. The results were used to question the importance of floodwater O2 concentrations above anoxia for submergence tolerance of rice.

A long-term field trial was conducted to determine yield trends in relation to nutrient uptake and efficiency in different rice-based cropping systems. The cropping systems had a significant effect on wet season rice yield when residues were not recycled but had no effect otherwise. Rice yield decreased after the first year of crop residue incorporation but increased every year thereafter. Rice yield was significantly affected by residual nutrients applied to dry season crops. The highest residual effect was observed in tomato and sweet pepper to which the highest nitrogen (N), phosphorus (P) and potassium (K) rates were applied. Maize, sweet pepper, and tomato responded well to NPK application, garlic had a low response and mungbean had no response. A relay crop served as a catch crop for excess nutrients and as shade to minimize sunscald effects for tomato and sweet pepper fruits.

A long-term field experiment was conducted to determine the effects of indigo (Indigofera tinctoria) on the productivity of rainfed lowland rice-based cropping systems: rice–tomato, rice–tobacco or soyabean, rice–maize, and rice–garlic. Indigo was grown as an intercrop during the dry season and incorporated as green manure for wet season rice. Dry season crop yields were generally not affected by the indigo intercrop but indigo green manure had a positive effect on rice yields. At the same level of nitrogen (N) inputs, indigo-N produced higher yields than urea-N which may be due to high inorganic N losses.

Many farmers are intensifying production systems by applying greater amounts of inorganic fertilizers, irrigation and pesticides, especially to cash crops. Such intensified systems, even though economically profitable in the short run, may not be sustainable. This paper analyses the economics and sustainability of an intensified rainfed rice-based system in Ilocos Norte, Philippines. Farmers use high levels of inorganic fertilizers for cash crops such as sweet pepper, garlic and tomato. Although these crops generate high levels of income, the high input systems may not be sustainable in the long run due to adverse on-site and off-site effects. Preliminary estimates of total factor productivity that include on-site effects only, display no clear time trend. However, negative externalities created by high nitrate contamination of groundwater and high rates of pesticide usage could make the system unsustainable by adversely affecting human health and the environment.

Under rainfed conditions in Tarlac, Central Luzon, Philippines, risk was characterized using field and household panel data for 46 rice farmers over the period 1990–93. Measures of risk at the field level were calculated using pooled time-series cross-sectional data. Field-level risk was found to be quite high, with the average coefficient of variation of yield being 45% and 30% for fields with low and moderate elevations respectively. Farmers applied less nitrogen in fields with higher yield variability, and adjusted the quantity of nitrogen in order to reduce losses in poor years and benefit from greater potential in good years. Farmers with higher levels of education had lower variability of rice income than farmers with a lower level of education. Non-crop income helped reduce total income variability, especially for farmers with high variability of crop income. Implications for technology design and policy improvements are derived from these results.