International Experience with the System of Rice

Intensification (SRI)

Rice Research and Training Institute

Kafr-el-Sheikh, April 8, 2010

Prof. Norman Uphoff, CIIFAD

Purpose: to share knowledge about a new system of crop

management• Not trying to sell SRI or to promote its use – rather would like to promote its evaluation• If SRI ideas and practices are found to be beneficial, this is open-access, public-domain information – available to anybody to use

• No patents, no royalties, no IPRs; not just rice

• SRI is not a technology – instead we refer to it as a system or a methodology• We think of SRI as is emerging paradigm for agricultural research and practice

Acc. to IRRI DG, Intl. Year of Rice, 2004

Rice sector needs in 21st century:

• Increased land productivity-- higher yield• Higher water productivity -- crop per drop• Technology accessible for the poor• Environmental friendliness • More pest- and disease-resistance • Tolerance of abiotic stresses (climate Δ) • Better grain quality for consumers• Greater profitability for farmers

SRI practices already can meet these needs:

• Higher yields -- 50-100% (or more?)

• Water reduction -- 25-50% (also rainfed)• Reduced input need - good for poor farmers• Little or no need for agrochemical inputs• Induced pest and disease resistance • Drought tolerance; little or no lodging• Better grain quality -- less chalkiness • Lower costs of production by 10-20% which leads to higher farmers’ income

SRI is many things:– SRI derives from a certain number

of INSIGHTS, based on experience – SRI can be explained in terms of

PRINCIPLES having scientific bases – SRI gets communicated to farmers

in terms of specific PRACTICES that improve the growing environment for their rice plants - at same time,

– SRI offers an alternative PARADIGM a different approach to agriculture - pointing toward post-modern agriculture

SRI is NOT A TECHNOLOGY SRI practices may look like a PACKAGE

or even like a RECIPE, but they are much better understood as a MENU

• Farmers are encouraged to use as many of the practices as possible, as well as possible

• Each practice contributes to higher yield as seen from the accumulating evidence

• But there is evidence of some synergy in effect among the practices – so that the best results come from using them together

SRI is NOT YET FINISHED -- Since SRI was empirically developed,

we are continually improving scientific understanding of SRI concepts/theory

-- Since SRI is farmer-centered, it is being modified, improved, extended

• There are now also rainfed versions of SRI and zero-till, direct-seed, raised-bed forms

• SRI ideas are extrapolated to other crops: wheat, sugar cane, millet, teff, beans, etc.

System of Finger Millet I ntensificationon lef t; regular management of improved

variety and of traditional variety on right

Liu Zhibin, Meishan, Sichuan province, China, standing in raised-bed, zero-till SRI field; measured yield 13.4 t/ha.In 2001, his SRI yield set provincial yield record: 16 t/ha

SRI was developed for smallholders in Madagascar, but it is relevant at all scales

- Fr. Henri de Laulanié arrived there from France in 1961 with some agricultural training- He started working with farmers to raise yields, without dependence on external inputs- In 1983-84 season, he learned the effects of transplanting young seedlings- In late 1980s, when fertilizer subsidies were removed, he switched over to compost

Fr. de Laulaniémaking field visit

Status of SRI: As of 1999

Known and practiced only in Madagascar

Before 1999: Madagascar1999-2000: China, Indonesia2000-01: Bangladesh, Cuba, Laos, Cambodia, Gambia, India, Nepal, Myanmar, Philippines, Sierra Leone, Sri Lanka, Thailand 2002-03: Benin, Guinea, Peru, Moz 2004-05: Senegal, Mali, Vietnam, Pakistan

2006: Burkina Faso, Bhutan, Iran, Iraq, Zambia2007: Afghanistan, Brazil 2008: Rwanda, Costa Rica, Ecuador, Egypt, Ghana2009: Malaysia Timor Leste2010: Kenya, DPRK, Panama? Solomon Islands?

2010: SRI benefits have now been validated in

39 countries of Asia, Africa, and Latin America

The Six Basic Ideas for SRI1. Transplant young seedlings to preserve their growth

potential -- but DIRECT SEEDING is now an option2. Avoid trauma to the roots -- transplant quickly and

shallow, not inverting root tips, which halts growth3. Give plants wider spacing -– one plant per hill and in

square pattern to achieve ‘edge effect’ everywhere4. Keep paddy soil moist but unflooded –- soil should be

mostly aerobic and not continuously saturated5. Actively aerate the soil as much as possible6. Enhance soil organic matter as much as possible

1+2+3 stimulate plant growth overall, while 4+5+6 enhance the growth of plants’ ROOTS and of soil BIOTA better PHENOTYPES

Cuban farmer with two plants of same variety

(VN 2084) and same age (52 DAP)

Single-seed SRI rice plantVariety: CiherangFertile tillers: 223Sampoerna CSR Program, Malang, E. Java, 2009

Additional benefits of SRI practice:• Reduced time to maturity,by 1-2 weeks, less exposure to pests and diseases, and to adverse climate; can replant sooner • Higher milling outturn – by about 15%

• Fewer broken grains, less chaff (unfilled grains)

•Human resource development for farmers through participatory approach – we want farmers to become better managers of their resources, experimenting, evaluating…• Diversification and modernization of smallholder agriculture; can adapt SRI to larger- scale production thru mechanization

Requirements/Constraints1. Water control to apply small amounts of water

reliably; may need drainage facilities2. Supply of biomass for making compost – but

can use fertilizer if compost is insufficient3. Crop protection may be necessary, although

usually more resistance to pests & diseases4. Mechanical weeder is desirable as this can

aerate the soil at same time it controls weeds5. Skill & motivation of farmers most important;

need to learn new practices; once techniques are mastered, SRI can become labor-saving

6. Support of experts? have faced opposition

SRI

0

50

100

150

200

250

300

IH H FH MR WR YRStage

Org

an d

ry w

eigh

t(g/

hill)

CK

I H H FH MR WR YR

Yellowleaf andsheathPanicle

Leaf

Sheath

Stem

47.9% 34.7%

Non-Flooding Rice Farming Technology in Irrigated Paddy FieldDr. Tao Longxing, China National Rice Research Institute, 2004

CHINA: Factorial trials by China National Rice Research Institute, 2004 and 2005

using two super-hybrid varieties --seeking to break ‘plateau’ limiting yields

Standard Rice Mgmt• 30-day seedlings• 20x20 cm spacing• Continuous

flooding• Fertilization:

– 100% chemical

New Rice Mgmt (SRI)

• 20-day seedlings• 30x30 cm spacing• Alternate wetting

and drying (AWD)• Fertilization:

– 50% chemical, – 50% organic

Average super-rice yields (kg/ha) with new rice

management (SRI) vs.standard rice management

at different plant densities ha-1

0100020003000400050006000700080009000

10000

150,000 180,000 210,000

NRMSRM

NRM plants are different phenotype, responding positively to less seed, less water, less fertilizer now being

wasted

2009 Report from Aga Khan

Foundation: Baghlan Province,

Afghanistan

2008: 6 farmers got SRI yields of 10.1 t/ha vs. 5.4 t/ha regular2009: 42 farmers got SRI yields of 9.3 t/ha vs. 5.6 t/ha regular

2nd year SRI farmers got 13.3 t/ha vs. 5.6 t/ha1st year SRI farmers got 8.7 t/ha vs. 5.5 t/ha

AFGHANISTAN: SRI field in Baghlan Province, supported by Aga Khan Foundation Natural Resource Management

program

SRI field at 30 days

SRI plant with 133 tillers @

72 days after transplanting

11.56 t/ha

IRAQ: Comparison trials, Al-Mishkhab Rice Research Station, Najaf

BHUTAN: Report on SRI in Deorali Geog, 2009

Sangay Dorji, Jr. Extension Agent, Deorali Georg, Dagana

SRI @ 25x25cm 9.5 t/ha SRI random spacing 6.0 t/ha

SRI @ 30x30cm 10.0 t/ha Standard practice 3.6 t/ha

MALI: SRI nursery in Timbuktu region – 8-day seedlings ready for transplanting

SRI transplanting in Timbuktu, Mali

MALI: Farmer in Timbuktu region

showing difference between regular

and SRI rice plants --

First SRI yields 8.98 t/ha (2007)

  SRI ControlFarmer Practice

Yield t/ha* 9.1 5.49 4.86Standard Error (SE) 0.24 0.27 0.18% Change compared to Control + 66 100 - 11% Change compared to Farmer Practice

+ 87 + 13 100

Number of Farmers

53 53 60

• * adjusted to 14% grain moisture content

MALI: Next year results - rice grain yield for SRI, control, and farmer-

practice plotsGoundam circle, Timbuktu region, 2008

Two Paradigms for Agriculture:

• GREEN REVOLUTION strategy was to:* Change the genetic potential of plants, and

* Increase the use of external inputs -- more water, more fertilizer and insecticides

• SRI (AGROECOLOGY) instead changes the management of plants, soil, water & nutrients:

* To promote the growth of root systems, and

* To increase the abundance and diversity of soil organisms to better enlist their benefits

The goal is to produce better PHENOTYPES

MADAGASCAR: Rice field grown with SRI methods

CAMBODIA: Rice plant

grown from a single seed in

Takeo province

NEPAL:Single rice

plant grownwith SRI methods, Morang district

IRAN: SRI roots and normal

(flooded) roots: note difference in color as well as size

INDONESIA:Rice plants ofsame age andsame variety

in Lombokprovince

Indonesia: Results of 9 seasons of on-farm comparative evaluations of SRI by Nippon Koei team, 2002-

06 • No. of trials: 12,133• Total area covered: 9,429.1 hectares• Ave. increase in yield: 78% (3.3 t/ha)• Reduction in water requirements: 40%• Reduction in fertilizer use: 50%• Reduction in costs of production: 20%

SRI LANKA: rice fields after three weeks without water;same rice variety, same irrigation system, same drought

-- conventional methods on left; SRI on right

VIETNAM: Dông Trù village,Hanoi province, after typhoon

has passed over --

SRI plant/field on left; conventional plant/field

on right

Nie Fu-qiu, Bu Tou village, Zhejiang

In 2004, SRI gave highest yield in province: 12 t/ha

In 2005, his SRI rice fields were hit by three typhoons – yet he was still able to harvest 11.15 tons/ha -- while other farmers’ fields were badly affected by the storm damage

In 2008, Nie used chemical fertilizer - and crop lodged

Irrigation method

Seed-ling age

Spacing(cm2)

Time to flowerin

g

Time to maturity

Plant lodging percentage

PartialComplet

eTotal

Inter-mittent

irrigation (AWDI)

1430x30 75 118 6.67 0 6.6730x18 74.67 118.67 40.00 6.67 46.67

2130x30 72.67 117.67 26.67 20 46.6730x18 74.33 117 13.33 13.33 26.67

Ordinary irrigation (continu-

ous flooding)

1430x30 73.33 122 16.67 33.33 50.0030x18 72 121 26.67 53.33 80.00

2130x30 72 120.67 20 76.67 96.6730x18 72.67 121 13.33 80 93.33

Time to flowering, maturity, and plant lodging percentage

as affected by AWDI and ordinary irrigation practice

combined with different age of seedlings and spacing

in Chiba, 2008 (Chapagain and Yamaji, 2009)

Incidence of Diseases and PestsVietnam National IPM Program: average of

data from on-farm trials in 8 provinces, 2005-06:

Spring season Summer season

SRIPlots

Farmer

Plots

Differ-ence

SRIPlots

Farmer

Plots

Differ-ence

Sheath blight

6.7%

18.1%

63.0% 5.2%

19.8%

73.7%

Leaf blight

-- -- -- 8.6%

36.3%

76.5%

Small leaf folder *

63.4 107.7 41.1% 61.8 122.3 49.5%

Brown plant hopper *

542 1,440 62.4% 545 3,214 83.0%

AVERAGE

55.5% 70.7%

* Insects/m2

Theory of Trophobiosis(F. Chaboussou, Healthy Crops, 2004)

deserves more attention and empirical evaluation than it has received to date

Its propositions are well supported by published literature over last 50 years -- and by long-standing observations about adverse effects of nitrogenous fertilizers and chlorinated pesticides

Theory does not support strictly ‘organic’ approach because nutrient amendments are approved where soil deficits exist

Theory of ‘Trophobiosis’

Explains incidence of pest and disease in terms of plants’ nutrition:

Nutrient imbalances and deficiencies lead to excesses of free amino acids in the plants’ sap and cells, not yet synthesized into proteins – and more simple sugars in sap and cytoplasm not incorporated into polysaccharides

This condition attracts and nourishes insects, bacteria, fungi, even viruses

Resistance to cold temperatures: Yield and meteorological data from ANGRAU, A.P.,

India

Period Period Mean max. Mean max. temp. temp. 00CC

Mean Mean min. min.

temp. temp. 00C C

No. of No. of sunshine hrssunshine hrs

1 – 151 – 15 NovNov 27.727.7 19.219.2 4.94.9

16–3016–30 Nov Nov 29.629.6 17.917.9 7.57.5

1 – 15 Dec1 – 15 Dec 29.129.1 14.614.6 8.68.6

16–31 Dec 16–31 Dec 28.128.1 12.212.2++ 8.68.6+ Sudden drop in minimum temp. for 5 days (16–21 Dec. = 9.2-9.9o C )

SeasonSeason Normal (t/ha)Normal (t/ha) SRI (t/ha)SRI (t/ha)

Kharif 2006Kharif 2006 0.21*0.21* 4.164.16

Rabi 2005-06Rabi 2005-06 2.25 2.25 3.473.47

* Low yield was due to cold injury (see below)

Measured Differences in Grain Quality

Conv. Methods SRI Methods Characteristic (3 spacings) (3 spacings) DifferenceChalky

kernels (%)

39.89 – 41.07 23.62 – 32.47 - 30.7%

General chalkiness (%)

6.74 – 7.17 1.02 – 4.04 - 65.7%

Milled rice outturn (%)

41.54 – 51.46 53.58 – 54.41 +16.1%

Head milled rice (%)

38.87 – 39.99 41.81 – 50.84 +17.5%

Paper by Prof. Ma Jun, Sichuan Agricultural University,

presented at 10th conference on “Theory and Practice forHigh-Quality, High-Yielding Rice in China,” Haerbin,

8/2004

NEPAL: Crop duration (from seed to seed) of different rice varieties with SRI (6.3 t/ha) vs.

conventional methods (3.1 t/ha) - 125 days vs. 141 days (16 days less)

Varieties (N = 412)

Conventional duration

SRI duration Difference

Bansdhan/Kanchhi

145 127 (117-144)

18 (28-11)

Mansuli 155 136 (126-146)

19 (29- 9)

Swarna 155 139 (126-150)

16 (29- 5)

Sugandha 120 106 (98-112) 14 (22- 8)

Radha 12 155 138 (125-144)

17 (30-11)

Barse 3017 135 118 17

Hardinath 1 120 107 (98-112) 13 (22- 8)

Barse 2014 135 127 (116-125)

8 (19-10)

Careful transplanting of single, young seedlings, widely spaced

SRI LANKA: Best use of transplanting methods

SRI LANKA: Soil-aerating hand weeder costs <$10

Effect of Active Soil Aeration

412 farmers in Morang district, Nepal, using SRI in monsoon season, 2005

SRI yield = 6.3 t/ha vs. control = 3.1 t/ha• Data show how WEEDINGS can raise yield

No. of No. of Average Rangeweedings farmers yield of yields

1 32 5.16 (3.6-7.6) 2 366 5.87 (3.5-11.0) 3 14 7.87 (5.85-10.4)

MechanicalWeedings

Farmers (N)

Area (ha)

Harvest(kg)

Yield (t/ha)

None 2 0.11 657 5.973One 8 0.62 3,741 7.723Two 27 3.54 26,102 7.373

Three 24 5.21 47,516 9.120Four 15 5.92 69,693 11.772

Impact of Weedings on Yield with SRI MethodsAmbatovaky, Madagascar, 1997-98

Mechanizationof weeding,

i.e.,soil aeration, is also possible

INDIA: Roller-marker devised by L. Reddy, East Godavari,AP, to save time in transplanting operations; yield in2003-04 rabi season was 16.2 t/ha paddy (dry weight)

SRI seedlings ready for transport to field, for mechanical transplanting

in COSTA RICA

Fig 1 Trasplantadora motorizada AP100 Yanmar

Mechanical transplanter

used in COSTA RICA

COSTA RICA - Mechanically transplanted SRI field 8 t/ha yield

PAKISTAN: Making raised beds for rice-growing

with adapted SRI methods on laser-leveled field

Mechanical transplanter for dropping seedlings into holes made by machine, Water is sprayed in hole after 10-day seedling is lput in,

adding compost.

Mechanical weeder set for spacing 9x9 inch (22.5x22.5 cm) – can give very good soil aeration

Rice crop at 71 days in Punjab, PakistanSeedlings planted into dry soil = 70% water

reductionAverage yield = 13 tons/ha (7 to 22

tons/ha)

Role of Soil BiotaBacteria and fungi perform many

services for crop (under aerobic conditions)

• Nutrient cycling and mobilization• Nitrogen fixation• Phosphorus solubilization• Water and nutrient acquisition• Competition with pathogens• Phytohormone production (auxins, etc.)• Induced systemic resistanceAlso previously unappreciated benefits

ENDOPHYTIC AZOSPIRILLUM, TI LLERING, AND RICE YIELDS WITH CULTIVATION

PRACTICES AND NUTRIENT AMENDMENTS Results of replicated trials at Anjomakely, Madagascar, 2001 (Andriankaja, 2002)

Azospirillum CLAY SOIL (Methods of cultivation)

in roots (103/mg)

Tillers/ plant Yield (t/ha)

Usual with no amendments 65 17 1.8 SRI with no amendments 1,100 45 6.1 SRI with NPK added 450 68 9.0 SRI methods with compost 1,400 78 10.5 LOAM SOIL SRI with no amendments 75 32 2.1 SRI methods with compost 2,000 47 6.6

Total bacteria Total diazotrophs

Microbial populations in rhizosphere soil in rice crop under different management at active tillering, panicle initiation and flowering (SRI = yellow; conventional = red)

[units are √ transformed values of population/gram of dry soil]

Phosphobacteria \ Azotobacter

Treatments Total microbes

(x105)

Azotobacter(x103)

Azospirillum(x103)

PSM(x104)

Conventional (T0) 2.3a 1.9a 0.9a 3.3a

Inorganic SRI (T1) 2.7a 2.2a 1.7ab 4.0a

Organic SRI (T2) 3.8b 3.7b 2.8bc 5.9b

Inorg. SRI + BF (T3)

4.8c 4..4b 3.3c 6.4bCFU = colony forming units PSM = Phosphate-solubilizing microbes BF = Bio-organic fertilizerValues with the different letters in a column are significantly different by LSD at the 0.05 level.Treatments: T0 = 20x20 cm spacing, 30 day seedlings, 6 seedlings/hill, 5 cm flooding depth of water, fertilized with inorganic NPK (250 kg urea, 200 kg SP-18, 100 kg KCl ha -1); T1, T2, T3 = All 30x30 cm spacing, 6-10 day seedlings, 1 seedling/hill, moist soil or intermittent irrigation, with different fertilization: T1 = same inorganic NPK as T0; T2 = 5 t ha-1 of organic fertilizer (compost); T3 = Inorganic NPK as in T0 + 300 kg ha-1 bioorganic fertilizer.

Total microbes and numbers of beneficial soil microbes (CFU g-1) under conventional and SRI rice cultivation methods, Tanjung Sari, Bogor district, Indonesia, Feb-Aug

2009 (Iswandi et al., 2009)

Ascending Migration of Endophytic Rhizobia,from Roots and Leaves, inside Rice Plants and Assessment of Benefits to Rice Growth

PhysiologyFeng Chi et al.,Applied and Envir. Microbiology 71 (2005),

7271-7278Rhizo-bium test strain

Total plant root

volume/pot (cm3)

Shoot dry weight/ pot (g)

Net photo-synthetic

rate (μmol-2 s-1)

Water utilization efficiency

Area (cm2) of flag leaf

Grain yield/ pot (g)

Ac-ORS571 210 ± 36A 63 ± 2A 16.42 ± 1.39A 3.62 ± 0.17BC 17.64 ± 4.94ABC 86 ± 5A

SM-1021 180 ± 26A 67 ± 5A 14.99 ± 1.64B 4.02 ± 0.19AB 20.03 ± 3.92A 86 ± 4A

SM-1002 168 ± 8AB 52 ± 4BC 13.70 ± 0.73B 4.15 ± 0.32A 19.58 ± 4.47AB 61 ± 4B

R1-2370 175 ± 23A 61 ± 8AB 13.85 ± 0.38B 3.36 ± 0.41C 18.98 ± 4.49AB 64 ± 9B

Mh-93 193 ± 16A 67 ± 4A 13.86 ± 0.76B 3.18 ± 0.25CD 16.79 ± 3.43BC 77 ± 5A

Control 130 ± 10B 47 ± 6C 10.23 ± 1.03C 2.77 ± 0.69D 15.24 ± 4.0C 51 ± 4C

Data are based on the average linear root and shoot growth of three symbiotic (dashed line) and three nonsymbiotic (solid line) plants.

Arrows indicate the times when root hair development started.

Ratio of root and shoot growth in symbiotic and nonsymbiotic rice plants

(symbiotic plants were inoculated with Fusarium culmorum)

Russell J. Rodriguez et al., ‘Symbiotic regulation of plant growth, development and reproduction,’ Communicative

and Integrative Biology, 2:3 (2009).

Growth of nonsymbiotic (on left) and symbiotic (on right) rice seedlings. On growth of endophyte (F. culmorum) and plant

inoculation procedures, see Rodriguez et al., Communicative and Integrative Biology, 2:3 (2009).

Extensions of SRI to Other Crops: Extensions of SRI to Other Crops: Uttarakhand / Himachal Pradesh, India Uttarakhand / Himachal Pradesh, India

Crop No. of Farmers

Area (ha)

Grain Yield (t/ha)

%Incr.

2006 Conv. SRI

Rajma 5 0.4 1.4 2.0 43Manduwa 5 0.4 1.8 2.4 33Wheat Research

Farm5.0 1.6 2.2 38

2007

Rajma 113 2.26 1.8 3.0 67Manduwa 43 0.8 1.5 2.4 60Wheat (Irrig.)

25 0.23 2.2 4.3 95

Wheat (Unirrig.)

25 0.09 1.6 2.6 63

Rajma (kidney beans)

Manduwa (millet)

Sugar cane grown with SRI methods (left) in Andhra Pradesh

Reported yields of 125-235 t/ha compared with usual 65 t/ha

ICRISAT-WWF Sugarcane

Initiative: at least 20% more cane

yield, with: • 30% reduction in water, and • 25% reduction in chemical inputs

‘The inspiration for putting this package together is from the successful approach of SRI – System of Rice Intensification.’

HIGH-TILLERING TRAIT IN TEFF WHEN TRANSPLANTED WITH WIDER SPACING

Report of Dr. Tareke Berhe, Senior Rice Advisor, Sasakawa Africa Association, Addis Ababa, from

experiments applying SRI methods for teff

1ST S.T.I. TRIALS, 2008 Duplication of Earlier Findings

VARIETYVARIETY SOWING SOWING METHODMETHOD

PELLETINGPELLETING YIELD YIELD (Kg/Ha)(Kg/Ha)

Cross 37Cross 37 BroadcastBroadcast NoneNone 1,0141,014

BroadcastBroadcast YesYes 483483

20 cm x 20 cm20 cm x 20 cm NoneNone 3,3903,39020 cm x 20 cm20 cm x 20 cm YesYes 5,1095,109

Cross 387Cross 387 BroadcastBroadcast NoneNone 1,1811,181

BroadcastBroadcast YesYes 1,0361,036

20 cm x 20 cm20 cm x 20 cm NoneNone 4,1424,14220 cm x 20 cm20 cm x 20 cm YesYes 4,3854,385

YIFRU ( 1998 ) M. Sc. THESIS

Reported yield of 4-5 tons/ha for

non-lodged teff vs.2-3 t/ha for lodged

teff

Even higher yields with addition of some macronutients (NPK) and micronutrients (Zn, Cu, S, Mg) – evaluations

continuing

SRI is pointing the way toward a paradigm shift? toward ‘post-modern agriculture’?• Less genocentric and more fundamentally biocentric• More interest in epigenetics• Re-focus biotechnology and bioengineering to utilize the benefits of biodiversity and ecological interactions• Less chemical-dependent and more energy-efficient• More oriented to health of humans and the environment•Intensification of production• Focus on greater factor productivity and sustainability

THANK YOU

• Can check out SRI website: http://ciifad.cornell.edu/sri/

• Email: ciifad@cornell.edu or ntu1@cornell.edu

How to “speed up the biological clock”

(adapted from Nemoto et al. 1995)

Shorter phyllochrons Longer phyllochrons• Higher temperatures > cold temperatures• Wider spacing > crowding of roots/canopy• More illumination > shading of plants• Ample nutrients in soil > nutrient deficits• Soil penetrability > compaction of soil• Sufficient moisture > drought conditions• Sufficient oxygen > hypoxic soil conditions

Root cross- sections of varieties:upland (left) and irrigated (right)

ORSTOM research (Puard et al. 1989)