1. What is Being Learned about the System of Rice Intensification (SRI)in China and Other Countries Norman Uphoff, CIIFAD CAWG-MOIST Seminar, Sept. 15, 2004

2. 3. Message: For Centuries, Even Millennia, We Have Been ABUSING and MISTREATING the Rice Plant

We have FLOODED it drowning its roots

We have CROWDED it inhibiting thegrowth potentialof its canopy and roots

Now we apply various FERTILIZERS and chemical BIOCIDES affecting thesoil biotawhich providevaluable servicesto plants:N fixation, P solubilization, protection against diseases and abiotic stresses, etc.

4. The System of Rice Intensification

Evolved in Madagascar over 20 years byFr. Henri de Laulani, S.J. working with farmers, observing, doing experiments, also having some luck in 1983-84 season

SRI is now spreading in countries around the world: positive results now seen in 21

SRI is a set ofprinciples and insightsthat when translated into certainpracticescanchange the growing environment of riceto get healthier, more productive plants

5. Canopy of an individual rice plant grown under SRI conditions; usually this variety (Swarna) is shy-tillering Andhra Pradesh, India, rabi season, 2003-04 6. Roots of a single rice plant (MTU 1071)grown at Agricultural Research Station Maruteru, AP, India, kharif 2003 7. SRI field in Sri Lanka -- yield of 13 t/ha with panicles having 400+ grains 8. CFA Camilo Cienfuegos, Cuba 14 t/ha -- Variety Los Palacios 9 9. SRI in Summary :

A set of principles/methods that getmore productivePHENOTYPES fromany existing GENOTYPEof rice.

SRIchanges the managementofplants, soil, water, and nutrientsto

(a) induce greater ROOT growth and

(b) nurturemore abundant and diversepopulations of SOIL BIOTA

10. Plant Physical Structure andLight Intensity Distributionat Heading Stage (CNRRI Research --Tao et al. 2002) 11. Dry Matter Accumulation between SRI and Control (CK) Practices(kg/ha) at Maturity (Zheng et al., SAAS, 2003) 12. Table 2. Different sizes of the leaf blade (cm) with SRI practices (Zheng et al., SAAS, 2003) 11.98 15.95 7.96 18.49 19.11 14.97 9.79 14.59 % 0.20 8.86 0.16 9.00 0.30 9.29 0.14 8.18 +/- 1.67 55.56 2.01 48.67 1.57 62.03 1.43 56.07 CK 1.87 64.41 2.17 57.67 1.87 71.32 1.57 64.25 SRI Width Length Width Length Width Length Width Length Average Flag leaf 2 ndleaf 3 rdleaf Prac-tice 13. Figure 1. Change of leaf area index (LAI) during growth cycle (Zheng et al., 2003) 14. Different P aradigmsof Production

The GREEN REVOLUTIONparadigm:

(a) Change thegenetic potentialof plants, and

(b) Increase the use ofexternal inputs-- more water, fertilizer, insecticides, etc.

SRI changes variousmanagement practices for plants, soil, water and nutrients, which:

(A) Promotethe growth of root systems , and

(B) Increase theabundance and diversityof

soil organisms ,and also

(C ) Reducewater use and costs of production

15. Greatest Benefit Is notYIELD

This can vary, often widely; for farmers,profitabilityis more important outcome

From societys perspective, what is most important isfactor productivity kg of rice perland, labor, capital, and water!

No question any longer ofwhetherSRI methods give higher yields/productivity but ratherhow to explainthese changes

For some things, we haveevidence ; for others,strong hypothesesfrom literature

16. What Are the Negatives?

Surprisingly few, but main one islabor intensity at least initially

This is receding as a constraint, mostlya problem for first few weeks or seasons

• Cambodian evaluation showed no increase (305 vs. 302 hrs/ha) -- and better timing

• IWMI showed labor productivity higher by 50-62%, with just partial use of SRI method

• Farmer innovation is helping to reduce labor requirements; more innovations will come

17. Roller-marker devised by Lakshmana Reddy, East Godavari, AP, India, to save time in transplanting operations; his yield in 2003-04 rabi season was 16.2 t/ha paddy (dry weight) 18. 4-row weeder designed by Gopal Swaminathan, Thanjavur, TN, India 19. Seeder Developed in Cuba 20. What Are the Negatives?

Less tractable iswater control although this can often be solved by infrastructure or organization SRI makes changespay

• Important for best results but may not bea barrier (R. Bunch report from Cambodia)

Requirement offarmer learning and skill is a benefit as well as a cost

Disadoptionhas been reported as a problem only in Madagascar(e.g. Cambodia)

Nematodescan be a problem(e.g., Thailand)

21. Report from China, 2004

Several SRI introductionsconcurrently:

Dec. 1998 seminars atCAU and NAU ; first NAU field trials in 1999 (9.2-10.5 t/ha)

Article on inILEIA journal(Dec. 1998) was read byCNRRIscientist, tried 1999

Director ofCNHRRDCattendedIRRI seminar, Feb. 1999; and/or got SRI paper from June 2000 seminar at Cornell

3-S systemwas developed separately in1991-96, field tested by 1999 byNEAU

22. Chinese Adaptations

Triangular systemof planting Liu Zhibin, Meishan, Sich. got16 t/ha and award from Prov. DOA

3-S system uses 45 d seedlings because of cold temperatures;single seedlings planted with wide spacing (10,000 plants/mou), less water, more organic matter; but no active soil aeration -- use herbicides

23. Normal 3-S 24. Seedlings are started at the end of winter in plastic greenhouses 25. Seedling for transplanting -- and resulting plant 26.

May 20-June 2

Transplanting 27. Wide Spacing of Plants

Average spacing 15 cm (13-17 cm) by 40 cm (37-43 cm), 1-2 per hill

28. Vegetative Growth Vigorous tillering 29. Vigorous roots 30. 3-S roots on right 31. . 131 3-S Field with Variety 131 32. 3-S has good grain quality 33. Effects of Different Treatments within 3-S System -- 9,691.5 26.0 90.0 92.0 450 Kongyu 131 +6.8% 10,351.5 25.6 78.0 108.0 480 Dongnong 9914 +24.0% 12,020.5 27.3 90.0 131.0 375 Xixuyan 1 +28.6% 12,471.0 29.0 85.0 136.0 371 Dongnong 423 VARIETIES + 28.5% 10,030.5 25.8 83.0 106.8 438.6 3-S METHOD -- 7,808.5 24.8 83.1 70.1 540.5 Control (CK) Change Yield (t/ha) 1000-grain wt (g) Seed set (%) Grains/ paniclePanicles/m 2 34. Effects of Different Treatments within 3-S System +3.2% 10,323.4 25.7 79.8 141.0 357.0 6-03-04 +20.1% 11,967.5 25.9 81.5 145.0 391.0 5-21-04 -- 9,960.8 25.4 79.6 138.0 357.0 5-07-04 Panicles/m 2 TRANSPL. DATE +15.4% 11,080.9 24.03 89.5 120.2 21.7 1/hill +6.0% 10,174.2 22.77 88.0 132.5 19.6 2/hill -- 9,601.4 22.09 82.0 109.2 23.2 3/hill Panicles/ hill PLANTS/ HILL + 5.8% 11,544.7 25.2 77.0 133.7 445.1 43x14 cm -- 10,913.1 25.1 77.5 127.5 440.1 36x17 cm Change Yield (t/ha) 1000-gr weight (g) Seed set (%) Grains/ panicle Panicles/m 2 DENSITY: SPACING 35. Other Chinese Adaptations

Plastic mulchwith an upland version of SRI in Sichuan: 8 t/ha, less labor: gives weed control, conserves moisture, also better soil temperature; useplastic trays

Intercropping potatoes, rape, or mushrooms are good alternating crops

In general:using 20-25d plants and herbicide for weed control, withsingle seedlings widely spaced ,reduced waterandmore organic matter ; these methods givehigher yields , often withless labor

36. Chinese Results, 2004

HeilongjiongProvince: 10 t/ha in 2004 -- 44,000 ha under 3-S system

GuizhouProvince: set high-altitude record in 2003 with SRI 12.9 t/ha

ZhejiangProvince, Tian Tai County: 10.8 t/ha in 2003; 11.5-12 t/ha in 2004 with 20 ha (300 mu), up to 13.5 t/ha

• Farmer experimentation is occurring

37. SRI demonstration fields in Tien Tai, Zhejiang, China 38. Prof. Zhu Defeng, CNRRI, in field visit to SRI fields in Tian Tai, Zhejiang 39. Nie Fuqiu, Bu Tou village, Tian Tai, Zhejiang, describing his experiments within SRI system 40. Evaluation of SRI by CAU Xinsheng Village, Dongxi Township, Jianyang County, August 2004

2003 7 farmers used SRI (SAAS)

2004 398 farmers used SRI (65%)

2003 SRI plot size average 0.07 mu

2004 SRI plot size average 0.99 mu

86.6% of SRI farmers (65/75) said they would expand their SRI area next yearor keep their whole rice area under SRI

41. Xinsheng Village, Dongxi Township [N = 75] (20% sample of all users) RICE YIELD (kg/mu) 2002 2003* 2004 Standard403.73297.88375.77 Methods SRI--439.87507.16 ----------------------------------------------------------- SRI Increase (%) +46.6%+34.8%*Drought year[Water saving/mu = 43.2%] 42. Other Results Reported, 2004 Sichuan Province 60+ trials showed 10.5 t/ha average vs. 7.5 t/ha usual (double usual increase with hybrid rice) SAU 11.75 t/ha;Leshan 12.1 t/ha (10 300 mu);Meishan 13.4 t/ha;SAASfield demonstration (observed) 11.64 t/ha Hunan Province 13.5 t/ha in field demonstration of CNHRRD (SRI) Yunnan Province 18 t/ha CNRRI trial20.4 t/ha certified by Dept of S&T/SAU 43. Liu Zhibin, Meishan Inst. of Science & Technology, in raised-bed,no-till SRI field with certified yield of 13.4 t/ha 44. Biomass applied to SRI plots in Meishan, China 45. MEASURED DIFFERENCES IN GRAIN QUALITY CharacteristicSRI (3 spacings)ConventionalDiff. Paper by Prof. Ma Jun, Sichuan Agricultural University, presented at 10th conference on Theory and Practice for High-Quality, High-Yielding Rice in China, Haerbin, 8/2004 + 17.5 38.87 - 39.99 41.81 - 50.84 Head milled rice (%) + 16.1 41.54 - 51.46 53.58 - 54.41 Milled rice outturn (%) - 65.7 6.74 - 7.17 1.02 - 4.04 General chalkiness (%) - 30.7 39.89 - 41.07 23.62 - 32.47 Chalky kernels (%) 46. SRI Is not a Niche Innovation

(Dobermann,Agricultural Systems , 2004)

Nor is itVoodoo Science[ Cold Fusion, UFOs]

(Cassman & Sinclair,ACSSA , 2004)

Sheehy et al. conclude inField Crops Research(2004):

[SRI] has no major role in improving

rice production generally

This is not a valid conclusion, though presentedas being more scientific than our conclusions

SRI is not finished yet still evolving, still improving can give seminar onexplanations , already supported or justifiably hypothesized

47. The SRI Effect Is Not Magic

It should and can be explainable in terms of relationships that are consistent with what is known about rice scientifically and about rices interactions with its environment (G x E interaction)

A number of possible explanations have surfaced in recent years; not all need to be correct, since there are so many

We think some combination of the following can account for SRI performance

48. 1 stExplanation? Above-Ground Environment

Create the edge effect for the whole field

Avoid edge effect only for measurement;promote it agronomically (triangle spacing)

Too-close spacing affectsphotosynthesiswithin canopy: measurements at AARD (Sukamandi, Indonesia) found that with normal spacing,lower leaveswere being subsidized by the upper leaves; wider spacing enableswhole plantto contribute

49. 2nd Explanation? Nitrogen Provision

Rice yields increased 40-60% whensame amountof Nprovided equallyin both NO 3 andNH 4forms vs. whenallN is provided as NH 4 (Kronzucker et al., 1998)

BNF increases greatly withalternated aerobic/anaerobic conditions (Magdoffand Bouldin,Plant and Soil , 1970)

Contributions ofprotozoato N supply

Also contributions fromendophytes

50. 51. 3rd Explanation?Phosphorus Solubilization

This nutrient is often limiting factor, but

Large amounts of P in soil (90-95%) are present in unavailable form

Alternating wetting and drying of soilincreased P in soil solution by 85-1900%compared with soils just wet or just dry(Turner and Haygarth,Nature , May 2001)

Aerobic bacteria acquire P from unavailable sources during dry phase; during wet phase theylyse and release P into the soil solution

52. 4 thExplanation?Mycorrhizal Fungi

90+% of terrestrial plants derive benefits from and even depend on mycorrhizal associations(infections)

Mycorrhizal hyphae extend into soil and expand volume accessible to the plant by10-100x ,acquiringwater, P and other nutrients ; they also provide protective/other services to plants

Flooded riceforgoes these benefits

53. 5 thExplanation? Phytohormones

Aerobic bacteria and fungi produceauxins, cytokinins, gibberellins , etc. in the rhizosphere

Huge literature has documented effects of microbially-produced phytohormones (e.g., Frankenberger and Arshad, 1995)

Root growth in SRI plants probably isnot due just to physiological processeswithin the plants --stimulated by aerobic microorganisms?Roots are key to SRI

54. Single Cambodian rice plant transplanted at 10 days 55. Cuba -- Variety VN 2084 (Bolito) -- 52DAP 56. Dry Matter Distribution of Roots in SRI and Conventionally-Grown Plants at Heading Stage(CNRRI research: Tao et al. 2002) Root dry weight (g) 57. Root Oxygenation Ability with SRIvs. Conventionally-Grown Rice Research done at Nanjing Agricultural University, Wuxianggeng 9 variety (Wang et al. 2002) 58. Table 13: Root Length Density (cm. cm -3 ) under SRI, Modern (SRA) and Conventional Practice (from Barison, 2002) Results from replicated on-station trials 0.06 0.13 0.36 1.19 1.28 4.11 Conventional practice 0.07 0.15 0.31 0.55 0.85 3.24 SRAwithout fertilization 0.09 0.18 0.34 0.65 0.99 3.73 SRAwith NPK and urea 0.20 0.25 0.32 0.57 0.71 3.33 SRI-- without compost 0.23 0.30 0.33 0.61 0.75 3.65 SRI--with compost 40-50 30-40 20-30 10-20 5-10 0-5 Soil layers (cm) Treatments 59. Figure 8: Linear regression relationship between N uptakeand grain yield for SRI andconventional methods,using QUEFTS modeling (from Barison, 2002)Results are from on-farm comparisons (N = 108) 60. Figure 9: Estimation of balanced N uptake for given a grain yield for rice plants withthe SRI and conventional systems,using QUEFTS modeling (same for P and K) (Barison, 2002) Results are from on-farm comparisons (N = 108) 61. What Are Problems for SRI?

Labor Requirements initially more labor-intensive -- 25-50%

As farmers gain skill & experience, this is reduced, and SRI can even becomelabor-saving

GTZ evaluation of SRI in Cambodia: no difference in labor requirements (305 vs. 302 hrs/ha) better timing

CEDAC evaluation of 120 farmers who have done SRI for 3 years: 55% said easier; only 18% said more difficult

62. Adjustable-width weeder designed by Hari R., Moramanga, Madagascar (from IRRI design) 63. Labor-Saving Methods ofCrop Establishment

Tray methods being developedin China, also in Cuba

Sowing/Thinning methods started in India and Sri Lanka broadcasting pregerminated seed (25 kg/ha) or young seedlings -- thenweed as usual , creating wide spacing with a square pattern(sacrifice seed for labor)

64. What Are Problems for SRI?

2.Water Control needed to get the best results with SRI methods

This constraint but can be reduced by investment in physical facilities or in organization and management

Most rice-growing countries will need toreduce the allocations of water for rice sectorin coming yrs

SRI can helpreduce water demand

65. Emerging Benefits of SRI?

1. Resistance toAbiotic Stressesclimate becoming more extreme and more unpredictable

Observed resistance todrought (Sri Lanka, several years) ,hurricane (Sichuan Sept. 2002) ,typhoon (AP, India Dec. 2003) ,cold spell (AP, India February 2004)

Resistance tolodgingdue to roots?

66. Two rice fields in Sri Lanka -- same variety, same irrigation system, andsame drought: conventional methods (left), SRI (right) 67. Emerging Benefits of SRI?

2. Resistance toPests and Diseases widely reported by farmers probably reflecting theprotective servicesof soil microorganisms

3. HigherMilling Outturn ~ 15%:SRI paddy raises outturn in Indiafrom 66 to 75%; in Cuba, from 60 to 68-71%; adds to paddy yield

Fewerunfilled grains(less chaff)

Fewerbroken grains(less shattering)

68. Emerging Benefits of SRI?

4. HigherNutritional Valueof Rice?

SRI can be organic rice that is free from agrochemical residues

Possibly SRI has higher nutritional quality in terms ofmicronutrients needs to be evaluated scientifically

Larger root system gives higher grain weight and greater grain densityalso great nutrient density?

69. Emerging Benefits of SRI?

5. Conservation of RiceBiodiversity ?

Highest SRI yields come with HYVs and hybrids (all SRI yields >15 t/ha)

Buttraditional/local varietiesrespond very well to SRI practice, can produce yields of 6-10 t/ha, and even more

Traditional rices receivehigher price

Higher SRI yields make them popular

Get anorganic premiumfor export?

70. 71. LESSCAN PRODUCEMORE

byutilizingbiological potentials & processes

Smaller, younger seedlings becomelarger, more productive mature plants

Fewer plants per hill and per m 2will givehigher yieldif used with other SRI practices

Half as much water producesmore rice because aerobic soil conditions are better

Greater outputis possiblewith use of

fewer or evenno external/chemical inputs

Changes in management practices givedifferent phenotypes fromrice genomes

72. SRI STILL RAISESMORE QUESTIONSTHAN WE HAVE ANSWERS FOR

This shouldplease scientistslot of interesting new work ahead

We are linking with researchers and practioners around the world

Two-pronged strategy: researchand practice proceeding in tandem-- walking on both legs as Mao advised

73. SRI Experience CouldHelp to Us to Improve21 stCentury Agriculture

Nurturing of roots and soil biota is relevant for much of agriculture

We need an agriculture that is

• Less thirsty-- better roots will help

• Less dependent on fossil-fuel energysources -- fertilizer, mechanization

• Less dependent on agrochemicals--for sake of soil & water quality, for health

74. To raise world rice productionby 60% by 2030, we will need totriple N fertilizer applications(Cassman et al., 1998)

Who believes this isfeasible ?

Who believes this isdesirable ?

Who believes this iseconomically possible ?

Who believes this isenvironmentally sustainable ?SRI shows alternative way

75. 76. 77. SRI Data from Sri Lanka

SRI Usual

Yields(tons/ha) 8.0 4.2+88%

Market price(Rs/ton) 1,500 1,300 +15%

Total cash cost(Rs/ha)18,000 22,000-18%

Gross returns(Rs/ha) 120,000 58,500+105%

Net profit(Rs/ha) 102,000 36,500+180%

Family labor earnings Increased with SRI

Water savings ~ 40-50%

Data from Dr. Aldas Janaiah, IRRI agric. economist, 1999-2002; now at Indira Gandhi Development Research Institute in Mumbai; based on interviews conducted with 30 SRI farmers in Sri Lanka, October, 2002

78. IWMI Data from Sri Lanka

IWMI Evaluation (Namara, Weligamage, Barker 2003)

60 SRI and 60 non-SRI farmers randomly selected:

YIELD-- increased by 50% on average(not doing full SRI)

WATER PRODUCTIVITY-- increased by 90%

COST OF PRODUCTION(Rs./kg) -- lower by 111-209% with family labor, 17-27%at standard wage rate

LABOR PRODUCTIVITY(kg/hr) -- up 50% in yala (dry) season, up 62% in maha (wet) season

PROFITABILITY-- increased by 83-206%, depending on the wage assumed (family labor vs. paid labor)

RISK REDUCTION-- conventional farmers had net losses in 28% of seasons, SRI farmers in only 4%

79. 80. SRI CONCEPTS CAN BE EXTENDED TOUPLAND PRODUCTION Results of trials (N=20) by Philippine NGO, Broader Initiatives for Negros Development,withAzucenalocal variety (4,000 m 2area) --usingmulchas main innovation, not young plants 81. ROOT SYSTEM PROMOTION

SRI is becoming referred to in India (AP) as the root revolution -- key factor

Roots benefit from wider plant spacing, aerated soil, more soil organic matter --from bothcompostandroot exudation

Roots are supported by more abundant and diversified populations ofsoil biota -- bacteria and viruses produce PGRs

Plants aretwo-way streets , coevolvedw/ microorganisms, dependent on them

82. Root Research Reported byDr. Ana Primavesi (1980)

Shoot and root growth of maize (in g)grown in hydroponic solutions (14 days), with varying nutrient concentrations

Shoot Root

100% concentration447

200% concentration347

2% concentration3323

2% concentration when4356changed every other day

83. Contribution of SOIL MICROBIAL PROCESSES

Microbial activity is known to be crucial factor in soil fertility

The microbial flora causes a large number of biochemical changes in the soil thatlargely determine the fertility of the soil. (DeDatta,1981, p. 60, emphasis added)

84. Bacteria, funguses, protozoa, amoeba, actinomycetes, etc.

Decompose organic matter , making nutrients available

Acquire nutrients otherwise unavailable to plant roots

Improve soil structure and health -- water retention, soil aggregation, aeration, pathogen control, etc.

85. 86. 87. Effect of Young Seedlings

@ AnjomakelyClay Soil Loam Soil

SS/20/3/NPK 3.00 2.04

SS/ 8 /3/NPK 7.16 3.89

SS/ 8 / 1 /NPK 8.13 4.36

AS / 8 /3/NPK 8.15 4.44

AS / 8 /3/ Comp 6.86 3.61

SS/ 8/ 1 / Comp 7.70 4.07

AS / 8 / 1 /NPK 8.77 5.00

AS / 8 / 1 / Comp 10.35 6.39

Note: All of these averages are for 6 replicated trials

88. Effects of SRI vs. Conventional Practices Comparing Varietal and Soil Differences