The document discusses the need to rethink rice selection practices and intensification to improve rice production systems, emphasizing that traditional methods relying on increased material inputs are becoming unsustainable. It highlights the System of Rice Intensification (SRI) as a more effective alternative that focuses on knowledge, management, and genetic potentials, resulting in higher yields with lower resources. The findings suggest that improved phenotypic expression through innovative management leads to more resilient rice plants and better adaptation to climate change.

1. Improving the Phenotypic Expression of
Rice Genotypes: Reasons to Rethink
Selection Practices and ‘Intensification’
for Rice Production Systems
Norman Uphoff, Cornell University, USA; Vasilia Fasoula, University of
Georgia, USA; Anas Iswandi, Institut Pertanian Bogor (IPB), Indonesia;
Amir Kassam, University of Reading, UK (presenter); and A.K. Thakur,
Directorate of Water Management, ICAR, India
2014 International Rice Congress, Bangkok
Section C02, Panel E: Improved varieties for
intensive production systems

2. ‘Intensification’ has different meanings --
The role of rice breeding will differ according to
the way that ‘intensification’ is understood
The most common idea of intensification
-- coming from Green Revolution experience --
depends on increasing material inputs used with
new rice varieties bred to be responsive to
the application of more inorganic fertilizers,
more water, and agrochemical protection
Question: Will this strategy be sufficient?
World rice production needs to double by 2050
(Ray et al., 2013)

3. Country paddy yields (t ha-1
), 1959-2011, 3-year averages
from FAO and USAID statistics (IRRI, 2014)
The gains from this paradigm have been decelerating as it
encounters diminishing returns. We need to exploit more fully
the genetic potentials that we have and/or can improve.
Countries*
1959-61
1969-71
1979-81
1989-91 1999-2000
2009-11
Total %
increase
Bangladesh 1.67 1.70 1.89 2.59 3.77 4.20 151%
Brazil 1.69 1.34 1.46 2.14 3.25 4.53 176%
China 2.03 3.30 4.28 5.62 6.32 6.60 225%
India 1.53 1.67 1.86 2.62 3.01 3.30 116%
Indonesia 1.93 2.38 3.53 4.33 4.38 4.36 126%
Myanmar 1.65 1.71 2.45 2.85 3.14 3.29 100%
Pakistan 1.36 2.24 2.41 2.32 2.95 3.28 141%
Philippines 1.21 1.65 2.23 2.79 3.10 3.64 200%
Thailand 1.65 1.93 1.85 2.10 2.60 2.83 71%
Vietnam 1.94 2.07 2.15 3.18 4.25 5.44 180%
Average yield 1.67 2.01 2.41 2.99 3.57 4.15 149%
% increase over
the decade
1960s
20.4%
1970s
19.4%
1980s
23.2%
1990s
19.3%
2000s
16.2%
*Producing 85% of the world’s rice

4. SRI is a different kind of intensification,
depending mostly on mental inputs
(knowledge) and different methods for
managing plants, soil, water and nutrients
SRI’s highest yields have come with improved
varieties
(tho it also helps unimproved varieties produce more)
•Using less inorganic fertilizer, to the extent that
organic materials are available and provided,
• Less water, by 30-50%, because no flooding, and
•Less need for agrochemical protection
More production with lower costs per hectare
gives farmers more profitability

5. Additional benefits from this kind of
agroecologically-grounded intensification:
More resistance to biotic and abiotic stresses:
•Drought-resistance – tolerate water stress
•Less lodging from wind and rain of storms
•Tolerance of more extreme temperatures
•Resistance to damage from pests and diseases
All are important for dealing with climate change;
also SRI makes net reductions in GHG emissions
Plus shortening of the crop cycle, by 5-15 days, and
higher milling outturn (>10%) because of
fewer unfilled grains and less breakage
So: more rice per unit time and per bag of paddy

6. Resistance to both biotic and abiotic stresses in East Java,
Indonesia: adjacent fields hit by both brown planthopper
(BPH) and by storm damage – the field on left was grown
with standard practices, while the field on right is organic SRI
Modern
improved
variety
(Ciherang)
– no yield
Traditional
aromatic
variety
(Sintanur)
- 8 t/ha

7. How are such effects possible? By eliciting
different, more productive, more resilient
phenotypes from given genotypes
Modification of management practices for
rice plants, soil, water and nutrients leads to
changes in plant morphology and physiology and
at same time to changes in the soil environment:
Pay more attention to E in equation: P = ƒ G + E + GxE
•Get larger, deeper, healthier root systems, and
•More abundant, diverse and active soil biota --
living around, on, and even inside the plants
Both contribute to better plant phenotypes

8. Effects of inoculation with Rhizobium leguminosarum bv. trifolii E11
on root architecture of two rice varieties: (a) Rootlets per plant;
(b) Cumulative root length (mm); (c) Root surface area (cm2
);
(d) Root biovolume (cm3
). Y. G. Yanni et al., Australian Journal of
Plant Physiology, 28, 845–870 (2001)
Indeed, there are positive interactions between
microbial populations and roots’ growth
Root systems support soil microbes, and v.v.

9. NEPAL:
Farmer with
a rice plant
grown from a
single seed
with SRI
methods
in Morang
district

10. LIBERIA:
Farmer with
rice plants of
same variety
and age: usual
methods on
left, and SRI
methods on
right in Grand
Gedee country

11. INDIA:
Farmer with
rice plants of
same variety
and age: usual
methods on
right, and SRI
methods on
left, grown in
Punjab state

12. CUBA: Two plants of the same age (52 DAS) and same variety
(VN 2084) -- different phenotypes from the same genotype

13. SRI
0
50
100
150
200
250
300
IH H FH MR WR YRStage
Organdryweight(g/hill)
I H H FH MR WR YR
CK Yellow leaf
and sheath
Panicle
Leaf
Sheath
Stem
47.9% 34.7%
CHINA: Research done at China National Rice Research Institute
by Dr. Tao Longxing, 2004 – same variety, different phenotypes

14. IRAQ: Pairs of trials at Al-Mishkhab Rice Research Station, Najaf, 2007,
comparing varieties with SRI management (on left) and RMP (on right)

15. INDONESIA: Stump
of a rice plant
(modern variety:
Ciherang cv) grown
from a single seed
using SRI methods
-- with 223 tillers and
massive root growth
Panda’an, E. Java, 2009

16. Data from comparative evaluations of SRI effects
on the physiology and morphology of rice plants,
conducted 2005-10 at ICAR’s Directorate of Water
Management in Bhubaneswar, India
The variety planted was Surendra (IET-12815),
130-135 days, usual yield 3.5-5.0 t ha-1
(DRD, 2006)
Published by Thakur et al. in Exper. Agric. (2010),
Paddy & Water Envir. (2011), Plant & Soil (2013)
SRI practices and RMP were taken from the
respective websites of SRI-Rice (Cornell) and
the Central Rice Research Institute (Cuttack)

17. Table 1: Effects of rice management practices on morphological
characteristics of roots, tillers, leaves, and canopy
Parameters
Management practices Increase
with SRISRI RMP LSD.05
Root growth parameters below-ground
Root depth (cm) 33.5 20.6 3.5 63%
Root dry weight (g hill-1
) 12.3 5.8 1.3 112%
Root dry weight (g m-2
) 306.9 291.8 NS 5%
Root volume (ml hill-1
) 53.6 19.1 4.9 111%
Root volume (ml m-2
) 1340.0 955.0 180.1 40%
Root length (cm hill-1
) 9402.5 4111.9 712.4 129%
Root density (cm-2
) 2.7 1.2 0.2 125%
Tillers, leaves and canopy structures above-ground
Plant height (cm) 124.2 101.4 8.1 22%
Tiller number hill-1
18.3 8.9 3.5 106%
Tiller number (m-2
) 450.1 441.2 NS 2%
Leaf number (hill-1
) 79.8 35.6 15.8 124%
Leaf number (m-2
) 1997.6 1766.5 229.4 13%
Leaf length (cm) 65.25 48.14 6.09 35%
Leaf widtha
(cm) 1.82 1.34 0.21 35%

18. Table 2: Effects of rice management practices on root
functions, physiological performance, N-uptake in rice
Parameters
Management practice Increase
with SRISRI RMP LSD.05
Amount of exudates (g hill-1
) 7.61 2.46 1.45 209%
Amount of exudates per m2
(g m-2
) 190.25 122.95 39.72 55%
Exudation rate per hill (g hill-1
h-1
) 0.32 0.10 0.06 220%
Exudation rate per m2
(g m-2
h-1
) 7.93 5.12 1.66 55%
Mean leaf elongation rate (cm day-1
) 5.97 4.45 0.21 36%
Chlorophyll a (mg g-1
FW) 2.35 1.68 0.14 40%
Chlorophyll b (mg g-1
FW) 1.02 0.90 0.07 13%
Total chlorophyll (mg g-1
FW) 3.37 2.58 0.11 30%
Chlorophyll a/b ratio 2.32 1.90 0.29 22%
Fv/Fm ratio 0.796 0.708 0.017 13%
Φ PS II 0.603 0.486 0.020 24%
Transpiration (m mol m-2
s-1
) 6.41 7.59 0.27 19%
Leaf temperature (°C) 34.48 33.09 NS 4%
Net photosynthetic rate (μ mol m-2
s-1
) 23.15 12.23 1.64 89%
N-uptake (kg N ha-1
) 77.4 51.0 8.6 52%

19. Table 3: Effects of rice management practices on
yield-contributing characteristics,
grain yield, straw weight, and harvest index
Parameters
Management practice Increase
with SRISRI RMP LSD.05
Panicle number hill-1
(ave.) 16.9 6.9 3.5 145%
Panicles (m-2
) 439.5 355.2 61.6 24%
Panicle length (cm) (ave.) 22.5 18.7 2.3 20%
Number of spikelets panicle-1
151.6 107.9 12.9 40%
Filled spikelets (%) 89.6 79.3 5.1 13%
1000-grain weight (g) 24.7 24.0 0.2 3%
Grain yield (t ha-1
) 6.51* 4.40* 0.26 48%
Straw weight (t ha-1
) 7.28 9.17 1.19 -21%
Harvest index 0.47 0.32 0.04 47%
* ICAR’s Directorate of Rice Research reports this variety’s yield as 3.5-5.0 t ha-1

20. Figure 1: Changes in crop growth rate (CGR) during the
vegetative stage of rice grown with SRI and RMP practices
Black circles = SRI management, and open circles = RMP
Vertical bars represent SEm ± (n=6)
0
10
20
30
40
50
60
30-40 40-50 50-60 60-70
Period (Days after germination)
CGR(gm-2
day-1
)

21. Figure 2: Changes in light interception by the canopy during
vegetative stage in rice grown with SRI practices and RMP
Black circles = SRI management, and open circles = RMP
Vertical bars represent SEm ± (n=6)
0
20
40
60
80
100
12 25 30 40 50 60 70
Days after seed germination
LightInterception(%)

22. Root Morphology and Physiology of Rice Plants
Cultivated under System of Rice Intensification (SRI)
Nurul Hidayati, Triadiati and Iswandi Anas, IPB, Indonesia
(Poster IRC2014-0616)
Culti-
vation
method
Root
aeren-
chyma
(%)
Stem
aeren-
chyma
area
(µm2
)
Total stem
aerenchyma
area (µm2
/
circumfer-
ence)
Conven-
tional
70.9b
53,597b
1,491,835b
SRI 45.1a
30,939a
83,5966a
Cultivation
method
Number of
root hairs
per mm2
Conventional 510.41b
SRI 816.50a
Conventional SRI

23. Rice physiology at vegetative, flowering, grain-filling, maturity phases
Orange lines = SRI; brown lines = conventional management.
Comparative research at IPB in Indonesia (Poster IRC2014-0595)

24. WHAT IS THE SYSTEM OF RICE INTENSIFICATION?
Simple Principles and Practices:
1. Undertake early, quick and healthy plant
establishment, minimizing transplant shock by
careful treatment of the plants’ roots
2. Reduce competition among plants through
wider spacing within and between hills
3. Improve the structure and functioning of the
soil with organic matter amendments
4. Maintain an aerobic soil environment for
plant growth through reduced or controlled water
applications & active soil aeration (aka weeding)

25. WHAT IMPLICATIONS FOR PLANT BREEDING?
• Selection of plants for breeding improved
lines should focus on the phenotypical
expression of individual plants, not on
group averages
• Plants being evaluated should be growing
with wide spacing, so their potentials for
great production are not diminished by
others’ competition
• This means that ‘honeycomb selection
designs’ are best suited to identifying
‘champions’ (Fasoulas and Fasoula, 1997),
able to screen large numbers of lines

26. Fig. 3. This replicated D-31 honeycomb design evaluates the plants of 31 sibling lines, grown in 26 rows with 23
plants per row. Each plant is in the center of a complete moving replicate, shown for two random plants of line
no.11 (gray circles). Plant Yield Index (PYI) measures plant yield devoid of confounding effects of soil heterogeneity.
Stability Index (SI) measures the stability of each sibling line by taking account of soil heterogeneity through
formation of the triangular grid that allocates plants uniformly across the whole field. The grid shown here is for
plants of line no. 11.

27. This strategy deals with a genetic relationship that
confounds conventional plant breeding strategies
• Genes for productivity are inversely related to
genes for competition (Fasoula, Euphytica,
1990; Fasoula & Fasoula, Plant Breeding
Review, 1997; Field Crops Research, 2002)
• Selection of ‘champions’ growing within
dense populations -- where genes for
competition are expressed -- will favor strong
competitors/ weak yielders over their
opposites
• The aim of breeding should be to select
density-neutral best producers, with all
plants in a population producing at their

28. COSTS OF PRODUCTION: TNAU study, Tamiraparani command area (N=100);
cost reduction with SRI system over conventional system = Rs. 2,369/ha (11 %);
labor input reduced by 8% (Thiyiagarajan, 2004 World Rice Research Congress)
Practices
Tractor
hours @
Rs.150/hr
Bullock
pair @
Rs.200/hr
Men’s
labour @
Rs.40/day
Women’s
labour @
Rs.40/day
Cost/ha (Rs.)
Con SRI Con SRI Con SRI Con SRI Con SRI
Nursery
preparation
1 - - - 6 3 0.5 5.5 2,110 681
Main field
preparation
7.5 7.5 2 2 12 12 - - 2,005 2,005
Manures &
fertilizers
- - - - 7 7 10 10 7,254 7,254
Transplanting - - - - 5 5 55 75 2,400 3,200
Weeding - - - - - 38 80 - 3,200 1,520
Irrigation - - - - 7.5 6 - - 300 240
Plant protection - - - - 2 2 2 2 660 660
Harvesting 1 1 - - 12.5 12.5 75 75 3,500 3,500
Total 9.5 8.5 2 2 52 85.5 222.5 167.5 21,429 19,060

29. Microbial populations in rice rhizosphere
Tamil Nadu Agricultural University research
Micro-
organisms
Standard
mgmt
SRI
mgmt
Difference
Total bacteria 88 x 106
105 x 106
1.2x
Azospirillum 8 x 105
31 x 105
3.9x
Azotobacter 39 x 103
66 x 103
1.7x
Phospho-
bacteria
33 x 103
59 x 103
1.8x
T. M. Thiyagarajan, WRRC presentation, Tsukuba, Japan, 2004

30. Total bacteria Total diazotrophs
Microbial populations in rice crops’ rhizosphere soil under conventional
crop management (red) and SRI management (yellow) at different
stages: active tillering, panicle initiation, and flowering. Units are √
transformed values of population/gram of dry soil (data from IPB)
Phosphobacteria
Azotobacter
0
10
20
30
40

31. Dehydrogenase activity (μg TPF) Urease activity (μg NH4-N))
Microbial activity in rice crops’ rhizosphere soil under conventional
crop management (red) and SRI management (yellow) at different
stages: active tillering, panicle initiation, and flowering. Units are √
transformed values of population/gram of dry soil per 24 h
Acid phosphate activity (μg p-
Nitrogenase activity (nano mol C2H4)

32. These results suggest the importance of
studying and understanding the
contributions that are made by microbes
living around, on and in plants =
symbiotic endophytes, which are
major components of the
plant-soil microbiome

33. “Ascending Migration of Endophytic Rhizobia, from
Roots and Leaves, inside Rice Plants and Assessment of
Benefits to Rice Growth Physiology”
Feng Chi et al., Applied and Envir. Microbiology 71: 7271-7278 (2005)
Rhizo-
bium
strain
Total plant
root
vol/pot
(cm3
)
± SE
Shoot dry
wt/pot
(g)
± SE
Net
photosyn-
thesis rate
(µmol of CO2
m-2
s-1
) ± SE
Water
utilization
efficiency
± SE
Grain
yield/pot
(g)
± SE
Ac-ORS
571
210
± 36A
63
± 2A
16.42
± 1.39A
3.63
± 0.17BC
86
± 5A
Sm-1021 180
± 26A
67
± 5A
14.99
± 1.64B
4.02
± 0.19AB
86
± 4A
Sm-1002 168
± 8AAB
52
± 4BC
13.70
± 0.73B
4.15
± 0.32A
61
± 4B
R1-2370 175
± 23A
61
± 8AB
13.85
± 0.38B
3.36
± 0.41C
64
± 9B
Mh-93 193
± 16A
67
± 4A
13.86
± 0.76B
3.18
± 0.25CD
77
± 5A
Control 130
B
47
C
10.23
C
2.77
D
51
C

34. “Proteomic analysis of rice seedlings infected by
Sinorhizobium meliloti 1021”
Feng Chi et al., Proteomics 10: 1861-1874 (2010)

35. Data are based on the average linear root and shoot growth of three
symbiotic (dashed line) and three non-symbiotic (solid line) plants.
Arrows indicate the times when root hair development started.
Ratio of root and shoot growth in symbiotic and
non-symbiotic rice plants -- seeds were inoculated
with the fungus Fusarium culmorum vs. controls
R. J. Rodriguez et al., ‘Symbiotic regulation of plant growth,
development and reproduction” Communicative
and Integrative Biology, 2:3 (2009).

36. Growth of nonsymbiotic (on left) and symbiotic (on right) rice seedlings.
On the growth of endophyte (F. culmorum) and plant inoculation procedures,
see Rodriguez et al., Communicative and Integrative Biology, 2:3 (2009).

37. More productive phenotypes also can give
higher water-use efficiency as measured by
the ratio of photosynthesis to transpiration
For each 1 millimol of water lost by transpiration:
3.6 micromols of CO2 are fixed in SRI plants,
1.6 micromols of CO2 are fixed in RMP plants
This becomes more important with climate change
and as water becomes a scarcer factor of production
“An assessment of physiological effects of the System of Rice
Intensification (SRI) compared with recommended rice cultivation
practices in India,” A.K. Thakur, N. Uphoff and E. Antony
Experimental Agriculture, 46(1), 77-98 (2010)

38. Economics, environmental vulnerabilities,
and climate change effects will require a
different kind of agriculture in 21st
century.
We need to REBIOLOGIZE AGRICULTURE
Fortunately, opportunities for a paradigm shift
are available; but they will require significant
changes in our crop and soil sciences
Work in microbiology, crop physiology, soil ecology,
and esp. epigenetics needs to become more central
to agricultural research and development

39. THANK YOU
Web page: http://sri.ciifad.cornell.edu/
Email: ntu1@cornell.edu [NTU-one]