Product Line 2.3. Rice varieties tolerant of abiotic stresses


Rice production in unfavorable environments is mostly constrained by abiotic stresses. These areas are commonly overpopulated, and are characterized by widespread and persistent rural poverty. About 30% of the 700 million people in absolute poverty (with income of < $1.25 per day) in Asia live in rainfed rice-growing areas in South Asia alone. The most serious abiotic stresses currently affecting rice production in Asia are drought, submergence, and salt stress, annually affecting about 23, 20, and 15 million ha, respectively. Low temperature adversely affects rice at high elevations and where rice is grown during the winter season in the subtropics, and heat stress is emerging as a serious threat to rice production as a consequence of climate change. The bulk of the rice produced in Africa is grown under rainfed conditions, accounting for more than 80% of the total rice cultivation. Stresses such as drought, salinity, submergence, low temperature, and iron toxicity are widespread in rice fields in Africa, contributing to persistent low rice yields. Iron toxicity and low soil fertility are common problems in Latin America, whereas low temperature is confined to southern Brazil, Uruguay, Argentina, and Chile. Considerable opportunities exist to at least double the yield in these areas through the use of stress-tolerant varieties. 

Products and activities

2.3.1. Drought-tolerant rice
All uplands risk water deficiency, and it is also a problem in rainfed lowland ecosystems. In the last decade, short-duration interspecific (NERICA) upland varieties have been widely deployed in Africa and they now occupy more than 300,000 ha. In addition, deep-rooting varieties are being used to develop drought-tolerant varieties with potential to extract water from deeper soil layers. Recent research at IRRI demonstrated potential for achieving an increase in yield of at least 1.0–1.5 t/ha under drought stress through breeding. Progress was also made in identifying major QTLs associated with yield under stress, and their effectiveness under field conditions was validated. Phenotyping techniques will be standardized to establish platforms for large-scale, precise measurements of yield and related traits under drought. NILs possessing drought-tolerance QTLs will be analyzed physiologically to unveil the interaction between these QTLs and facilitate their effective use in breeding. A wide range of genetic resources (O. sativa and other Oryza species indigenous in Africa) will be used for the development of drought-tolerant varieties.

2.3.2. Submergence-tolerant and other flood-tolerant rice
Rice is sensitive to flooding during germination, which hinders direct seeding in rainfed areas, and also during the vegetative stage when completely submerged. Stagnant partial flooding of 20–50 cm for most of the season also affects considerable areas, estimated at more than 5 million ha in India and Bangladesh alone. The SUB1 gene, which confers an advantage of 1–3 tons of grain yield following flooding for 10–15 days, will form the basis of tolerance of submergence in all breeding materials. This will be introduced into a wide range of genetic backgrounds to develop more tolerant varieties that are also adapted to longer-term stagnant flooding. Lines tolerant of anaerobic germination will be developed from the best sources identified, and through the use of marker-assisted selection. Tolerance traits of all submergence types will be characterized at the physiological level and the best donors for breeding will be identified and used in crosses. The SUB1 gene that confers submergence tolerance is now being transferred into African mega-varieties, using IRRI donors. The improved mega-varieties, in terms of submergence tolerance, will be evaluated on a large scale in multilocation trials at “hotspots” in Africa.

2.3.3. Improved varieties tolerant of salt stress and other problem soils
Poor soils with excess salt or deficiency in certain plant nutrients limit rice productivity in most rainfed rice areas, and several million ha of land suited to rice production in Asia and Africa are currently unexploited because of salinity and other related soil problems. Rice is suitable for reclaiming these soils because it thrives well under flooding, and has high potential for genetic manipulation. Rice productivity in salt-affected areas is very low, <1.5 t/ha, but this can reasonably be raised by at least 2 t/ha. Saltol and other QTLs for seedling-stage tolerance and one for the reproductive stage will be targeted to develop varieties tolerant at both stages, and the physiology and genetics of tolerance at both stages will be advanced. Nutritional imbalances such as P and Zn deficiency and Fe and Al toxicity are widespread in most rice production areas in Asia, Latin America, and Africa. Donors for tolerance of these soil problems will be identified and physiologically and genetically characterized, and the major QTLs identified and used in breeding. Iron toxicity is a widespread growth constraint in lowland rice in Africa. Several highly tolerant varieties/lines such as Suakoko 8 (O. sativa) and CG 14 (O. glaberrima) have been identified and some improved varieties have been released by AfricaRice that are tolerant of iron toxicity, including both intra- and interspecifics. At AfricaRice, breeding is under way to validate tolerance of iron toxicity and to identify new QTLs for use in MAS.

2.3.4. Varieties tolerant of cold or hot temperatures 
High temperature will become an increasing problem because of climate change. In the Sahel region of Africa, temperatures above 40 °C are experienced quite often during rice cultivation periods. Heat stress causes high sterility, leaf yellowing, and accelerated development leading to low yield potential in sensitive rice varieties. Rice plants are most sensitive at the flowering and ripening stages. Both yield and grain quality are adversely affected. Donors for tolerance of high temperature are being identified by screening improved and traditional rice varieties. O. glaberrima could be a useful genetic source since it has a habit of early-morning flowering and high transpiration with sufficient water, both of which are convenient traits for avoiding heat stress. These donors will be used in a crossing program to incorporate tolerance of high temperature into elite cultivars suitable for different growing environments. QTL mapping will facilitate the use of marker-assisted selection in developing improved heat-tolerant cultivars. Segregating populations from IRRI will be evaluated and selected at CIAT. Cold-tolerant cultivars are needed in both temperate regions and high-elevation tropical areas. Cold stress can be experienced during the whole crop cycle and thus tolerance is needed at both the seedling and reproductive stage. New genes and QTLs are being mapped that confer tolerance of low temperature at different growth stages. At AfricaRice, breeding for cold tolerance is being done for indica-type varieties in the Sahel zone and for japonica types in the highlands. However, for both ecologies, indica-type grain is mainly preferred by consumers and this is being integrated into the breeding objectives. In Latin America, cold-tolerant cultivars are needed both in the Southern Cone (Brazil, Uruguay, Argentina, and Chile) temperate regions and in some high-elevation tropical areas.

Uptake and impact pathway

The next users are rice breeders working in GRiSP, NARES breeding programs, and the private sector. Intermediate users are seed producers and distributors, including the public sector, private companies, and NGOs, and final users are farmers. It is assumed that there will be assured funding from donors and national governments to produce and deliver the required amount of seed to farmers. Government policies should facilitate the release of stress-tolerant cultivars. For instance, varieties released in one country in West Africa should automatically qualify for release in other ECOWAS member countries for the same ecology. For the unfavorable rice environments of Asia, CURE will play a key role to link improved germplasm with appropriate management practices and cropping systems. Germplasm sources should be available from theme 1 and management practices for unfavorable areas will be developed in theme 3, requiring close linkage. Participatory varietal selection (PVS) is a method that is necessary for the effective evaluation of improved germplasm with farmers, and participation of social scientists is essential. Seed production and dissemination will be handled by linkage with theme 6. Linkages with nationally and internationally funded development projects will facilitate the delivery of improved varieties to farmers. The Africa Rice Breeding Task Force, revitalized in 2010 with Japan’s support, will greatly accelerate region-wide varietal development/evaluation, dissemination, and harmonization of varietal nomination and release systems.

Financing strategy

The research is currently funded by several restricted grants that need to continue and expand under the new funding windows of the Consortium, including BMGF-STRASA (products 2.3.1–2.3.4), USAID-CSISA (2.3.4), the Japan Rice Breeding project, and other projects from BMZ, GCP, and RISOCAS. Work on low-temperature tolerance is funded largely through a project with RDA-Korea. National partners are contributing through in-kind support and funding for seed production activities. Additional funding on climate change activities should include high-temperature stress and tolerance of stagnant flooding and salinity (delta areas). In Latin America, work on low-temperature tolerance is funded largely through FONTAGRO; new funding will be required for work on high-temperature stress and to support more work on breaking the yield barrier.

 Box 10. Impact example for product 2.3.2: Submergence-tolerant rice

SUB1 is a major quantitative trait locus (QTL) that confers submergence tolerance in rice. The impact pathway of SUB1 provides an example of how new varieties are being developed through modern breeding tools in PL 2.3, with ensuing unsurpassed impacts in farmers’ fields (see table on next page). It involved the effective use of natural variation for germplasm improvement and efficient delivery methods in accelerating impacts.

Farmers in flash-flood areas have been using local landraces for decades because they mostly possess some tolerance of prevailing abiotic stresses, but with low productivity and poor quality. Collection and evaluation of this material from flood-prone areas identified FR13A from Orissa, India, as the best source of tolerance. Conventional methods to develop tolerant varieties made limited progress over a few decades. The success in genetic mapping of SUB1 has enabled marker-assisted breeding of high-yielding rice varieties capable of enduring complete submergence for up to 2 weeks, with a yield advantage of 1–3 t/ha demonstrated in farmers’ fields following complete inundation for variable durations. The triumphs of this technology prompted considerable interest and support from NARES. The first breeding line, Swarna-Sub1, was developed in 2006, and, by 2009, a few varieties were already formally commercialized in South and Southeast Asia.

Considerable efforts are still needed to transfer SUB1 into more varieties suitable for areas prone to flash-floods in Asia, Africa, and Latin America. In Asia alone, more than 20 million ha can benefit from this technology. Further improvements are needed to develop more resilient varieties for flood-prone areas, and to cope with the changing climate. Additional QTLs need to be identified to increase tolerance to at least 20 days of complete submergence, and, in most areas, tolerance of partial stagnant flooding is needed as a water level of up to 50 cm can prevail after flash-floods. Direct seeding in flood-prone areas will be possible only pending improvements in tolerance of flooding during germination and early crop establishment. Developing more durable varieties for flood-prone areas will thus require the combination of these tolerance traits in high-yielding genotypes possessing other agronomic and quality traits desired in particular target regions. Seeds of these varieties need to be delivered to farmers via effective seed systems, together with packages of adjusted management options to fully exploit their potential for better and sustained system productivity. A concerted global effort of numerous partners such as that being built in GRiSP is needed, involving CGIAR centers (IRRI, AfricaRice, CIAT), advanced research institutes, NARES, GOs and NGOs, the private sector, and farmers.

Past and projected research to impact pathway of submergence-tolerant rice (with the SUB1 gene)



1985 -90








Tolerant landraces collected

Gene-bank screening

Donor (FR13A) identified

Breeding combining tolerance and dwarf phenotype

IR49830-7 developed

SUB1 mapped to chr. 9 (70% phenotype)



Fine-mapping advanced

Markers developed for MABC

Cross of Swarna with IR49830 (2002)

MABC developed for SUB1

SUB1-A gene identified

Precise markers developed (2006)

Swarna-Sub1 developed through MABC (2006)

Six Sub1 varieties developed (2008)

SUB1 advantages quantified in farmers’ fields (1–3 t/ha after submergence)

SUB1 combined with one or more QTLs to provide protection for approx. 20 d of submergence

New varieties combining SUB1, AG, and tolerance of stagnant flooding available

50 Sub1 varieties developed

20 varieties combining AG and tolerance of submergence and stagnant flooding developed

Research partners






IRRI, UCD, UCR, CRRI, NDUAT, BRRI, PhilRice; other NARES from S and SE Asia

IRRI, UCD, UCR, AfricaRice, CIAT, NARES in Asia, Africa, and Latin America

IRRI, UCD, UCR, AfricaRice, CIAT, NARES in Asia, Africa, and Latin America

Delivery partners






AfricaRice, African NARES, ACIAR, CRRI, NDUAT, BRRI, PhilRice, BRAC, other NARES, GOs and NGOs in S & SE Asia

AfricaRice, CIAT, NARES, GOs and NGOs in Asia, Africa, and Latin America

AfricaRice, CIAT, NARES, GOs and NGOs in Asia, Africa, and Latin America

Delivery strategies/ milestones






Provision of seeds to NARES

PVS trials and extension

Facilitating release

Seeds via effective seed systems (GOs, NGOs, private sector)

Network efforts (INGER, shuttle breeding, CURE, STRASA, MoFA)

Provision of seeds to NARES

PVS trials and extension

Facilitating release

Seeds via effective seed systems (GOs, NGOs, private sector)

Provision of seeds to NARES

PVS trials and extension

Facilitating release

Seeds via effective seed systems (GOs, NGOs, private sector)

Adoption levels






Swarna-Sub1 released in India & Indonesia (2009)

Other Sub1 varieties released in Bangladesh (2), Philippines (1), and Indonesia (3) in 2010

Over 200 tons of seeds of Sub1 varieties distributed

Sub1 varieties cover over 0.1 million ha in Asia

Submergence-tolerant varieties released in 5 countries in Asia and 5 in Africa/Latin America

Yield losses from submergence reduced by 40% on 20% of flood-prone areas in Asia

Tolerant high-yielding varieties released in flood-prone areas of all countries in Asia, Africa, and Latin America

Tolerant varieties grown on 12 million ha of flood-prone areas with 40% reduction in yield losses

Partners: Details contained in the GRiSP partners database. MoFA: a project supported by the Japanese Ministry of Foreign Affairs (2007-09). AG = anaerobic germination.