A-006 Practice Of Sri In Sri Lanka

The practice and effects of the System of Rice Intensification (SRI) in Sri Lanka

Regassa Namara1, Deborah Bossio2, Parakrama Weligamage3 and Indika Herath4

1

Corresponding author, Economist, International Water Management Institute, PMB, CT 112,

Cantonments, Accra, Ghana. Tel: 233-21-784753/4, Fax: 233-21-784752, e-mail: [email protected] 2

Principal Researcher, International Water Management Institute, P.o.Box 2075, Colombo, Sri Lanka.

Tel: 94-11 2787404, Fax: 94-11 2786854. E-mail: [email protected] 3

Researcher, International Water Management Institute, P.O.Box 2075, Colombo, Sri Lanka. Tel: 94-11

2787404, Fax: 94-11 2786854, e-mail: [email protected] 4

Post Graduate Student, Post Graduate Institute of Agriculture, University of Peradenya, Sri Lanka. Formatted: English (U.S.)

Tel: 0812395200, Fax: 0812388041, e-mail: [email protected]

Field Code Changed Formatted: English (U.S.) Formatted: English (U.S.)

1

Abstract In Sri-Lanka, rice is grown under conditions of sub-optimal water and land availability. Thus, innovations such as the System of Rice Intensification (SRI) that can increase productivity and save resources are needed. The objective of this study was to understand how SRI was implemented on farms in Sri-Lanka, and the consequences of changes in practices on: 1) input utilisation, 2) agronomic traits of rice, and 3) soil chemical properties. It was found that SRI farmers had made significant changes in their production systems: irrigations, seeding rates and herbicide usage were reduced by 24%, 85% and 95% respectively, and plant spacing was increased by 60%. Total inputs of nitrogen, phosphorus, and potassium were similar across SRI and conventional plots, but the source of nutrients was different. Yields were variable, but significantly higher on SRI farms, soil available potassium and phosphorus were increased, and SRI plants exhibited tolerance to low moisture stress. Keywords: Agronomic analyses, Soil Analyses, Water Management, Rice

1 1.1

Introduction Background

The South Asian region has been transformed from a state of severe food shortages and starvation of large numbers of its people during 1960s, mainly due to the poor productivity of two of its staple cereals (Rice and wheat) to a state of surplus production. The main factors behind this transformation process, often characterized as a “green revolution,” were introduction of short-stature, fertilizer-responsive, lodgingand disease- resistant and high-yielding varieties; investments in irrigation infrastructure; increased use of chemical fertilizers, herbicides, insecticides and fungicides, and government support through extension and micro-credit provisions (Singh, 1990; Ellis 1993). This ‘green revolution,’ or conventional system of production intensification, had negative social and environmental externalities (Vandana, 1991). Many smallholder farmers have suffered from decreased grain prices that resulted from the massive increase in production (Evenson and Gollin, 2003). Market price declines combined with increases in the relative prices of inputs required by the conventional intensification process have resulted in a very small return to investment for farmers. Thus economies of scale, and reliable access to large quantities of inputs including water, fertilizers and pesticides, play an important part in farm viability, and many smallholder farmers have not benefited from these decades of investment in agricultural development to the extent

1

expected (Lipton and Longhurst, 1989). Surplus grain production has in fact become a national problem; governments in India hold buffer stocks of wheat and rice at high expense, in order to maintain minimum prices (Roy, 2003). It is now widely accepted that although the ‘green revolution’ contributed greatly to global famine mitigation, the benefits of this system of productivity intensification have not been shared equally among farmers or regions (IFPRI, 2002). Negative environmental externalities of conventional intensification systems are now also widely appreciated. Salinization and waterlogging affects 42 million ha and 4.6 million ha of agricultural lands, respectively in south Asia1 (FAO, 1994), and pollution from agricultural chemicals threatens drinking water sources and degrades wetlands and other aquatic habitats (FAO, 2002).

Despite global surpluses, in many local cases in developing countries increasing production is still desirable on a local level primarily for food security in poor rural areas, but also to provide food for growing cities, and export markets. Technologies that can lower costs, are favourable to the environment, and save resources such as water, are needed to provide for these various needs in the context of social equity and poverty alleviation concerns. Often, however, further increases in the production of rice depend on intensification on existing rice lands, using existing or less water resources. In Sri Lanka, as in much of South Asia, most land resources suited to the production of rice have already been exploited, and most of the readily manageable water resources have been developed to irrigate paddy fields. Water availability is a growing constraint for paddy rice cultivation, as competition for other uses of the water increases (Barker et al., 2001). Although the dominant practice in rice production in Sri Lanka is flooded irrigation, 28.2% of the rice is grown under rainfed conditions with suboptimal water availability (Dhanapala, 2000). Paddy crop failure due to drought in Sri Lanka occurs for 2 to 3 years out of 20, and yield decrease due to water shortage is more common (Tennakoon, 1986). This situation is not unique to Sri Lanka. Worldwide 37.4% of the paddy rice is grown under low land and upland rainfed conditions, where water supplies are either not adequate today, or are predicted to be inadequate by 2025, and yields suffer from either regular or intermittent water shortages (IRRI, 1997). Thus in Sri Lanka, as elsewhere in the world, technologies or production systems that can stabilize or increase production and save resources such as water are needed. Many alternative production intensification systems are available. These include: low external-input sustainable agriculture, organic farming, ecological farming, intermitent irrigation, alternate wetting and drying, and aerobic rice cultivation (Barker et al., 2001). The system of rice intensification (SRI) shares one or more of the aspects of these methods of production, but very little

1

The South Asian countries include India, Pakistan, Sri Lanka, Bangladesh, Bhutan, Nepal, Afghanistan and Iran

2

is known about how the system is actually implemented on farms, and its effects on resouce use and condition.

Following the reports of its dramatic yield and water productivity advantages, SRI has recently generated interest and discussions among researchers, development practitioners and policy makers. These discussions have often resulted in polarised views with one group advocating for the wider dissemination of the practice and another group questioning the plausibility of the reported advantages2. This paper will not indulge in this debate. Rather the objective of this study was to understand how SRI is actually implemented on farms in Sri Lanka, and the consequences of the change in practices on: 1) input utilisation, 2) selected agronomic traits of rice, and 3) soil chemical properties. Results are discussed with reference to country-wide trends in rice production practices, and the resourse conservation and environmental implications of SRI. The study incorporated a combination of socioeconomic survey data, agronomic and soils analysis, and secondary data sources.

1.2

System of Rice Intensification

The system of rice intensification was inductively developed in Madagascar by Fr. De Laulanie, a French priest and agriculturist through working with farmers (Uphoff et al., 2002). Many have given definitions and descriptions of SRI3. All of these definitions underline the importance of conceptualizing SRI as a system rather than as technology because it is not a fixed set of practices; it is rather a system of production based on certain core principles with the possibility of adjusting the exact technical components based on the prevailing biophysical and socio-economic realities of an area. SRI practices therefore, are still evolving with the goal of improving sustainability and factor productivity of land, labour, and water. The main elements of SRI are (1) planting method, (2) soil fertility management, (3) weed control, and (4) water (irrigation) management (table 1). It is recommended that these components be tested and varied according to local conditions rather than simply adopted (Stoop and Kassam, 2005). Table 1. The core elements of the System of Rice Intensification

2 3

Elements

Specific components

Description

Planting method

Spacing of seedlings

25cm by 25cm or more

Age of seedlings at transplanting

8-15 days after germination

For an interesting debate on SRI see Stoop and Kassam, Sheey et al. 2004, Sinclair and Cassman 2004 For descriptions and definitions of SRI see http://ciifad.cornell.edu/sri/index.html

3

Soil fertility management

Weed control

Number of seedlings

1 to 2 seedlings per hill on square grid

Duration from uprooting to planting

15-30 minutes

Organic matter

More use of organic matter

Chemical fertilizer

Chemical fertilizer use may be gradually avoided

Mechanical or hand weeding

Weeding is best done with mechanical weeder

Herbicide

Herbicide use may preferably be avoided

Irrigation (water) Management

2 2.1

Alternate Wetting and Drying

Apply small quantity of water daily. Flood and dry the field for alternating periods of 3-6 days

Materials and methods Location and socio-economic survey

The two study locations, Krunegala and Rathnapura districts, were purposively selected based on the prevalence of SRI farmers (see figure 1). Rathnapura district is predominantly a wet zone area, while south-eastern parts of it, where the specific study location or Kalthota irrigation scheme is located, fall within the intermediate and dry zones. Kalthota irrigation system is a river diversion system in the Walawe River, one of the major rivers in the country. Today the systems irrigate 1000 ha of land through two conveyance channels, situated on both sides of the river. The left bank irrigates 128 ha of paddy lands in three tracts while the right bank canal irrigates 728 ha situated in seven tracts. Kalthota farmers had the freedom of using the flow of the Walawe River until 1990 as there were no major upstream developments to divert water from the original river course. Construction of a reservoir upstream as the storage facility for a 120 MW hydropower generation complex affected the flow of water at the diversion point as the discharge after power generation is at the point below the water uptake to the system. Since this intervention, the Kalthota irrigation system is experiencing a shortage of water. SRI appeared to be a potential water saver so Ceylon Electricity Board took steps in promoting SRI among farmers in the irrigation system.

Fig. 1 Study locations

4

5

Krunegala district is situated in the north-western zone part of Sri Lanka and is a major paddy-producing area in the country. It is the third largest district in terms of land area. About 75 percent of the district falls within the intermediate zone and southern part falls within the wet zone while its northern part in the dry zone. The specific study locations in the district, i.e., Maharachchimulla and Maho-Rambe, are situated within the intermediate low country. In these locations, the Ministry of Samurdhi, responsible for poverty alleviation took a special interest in promoting SRI through farmer training programs.

A two stage stratified random sampling design was used to select 120 farmers in total (i.e., 60 each from the two study locations and 30 each from SRI and Non-SRI farmers per location). In Rathnapura (Kalthota irrigation scheme) we stratified SRI farmers into two based on their location relative to the irrigation scheme (namely left and right bank). Then a simple random sampling design was used to select 15 SRI farmers from each bank. The non-SRI farmers were selected in two steps. First, two Farmer Organizations from the left bank and four from the right bank were selected. Then in the second stage seven to eight farmers from each Farmer Organization were selected, making the total 30 non-SRI farmers.

In Kurunegala, SRI farmers were found dispersed over substantial areas of the district. To economize survey logistics and time farmers were selected from two locations in which 68% of all practicing SRI farmers were found; one to the northeast of Kurunegala (Maho-Rambe) and the other towards the southwest of the town (Maharachchimulla). A separate list of SRI farmers was prepared for each location, which then served as a sampling frame, and 15 SRI farmers each from Maho-Rambe and Maharachchimulla were selected. For the non SRI sample, two villages or Farmer Organizations from each location were selected to represent the dominant type of water regime of paddy farming in the respective area. These were Dagama (from Hakwatuna major irrigation scheme) and Ponnilawa Maha Wewa (15 ha minor tank scheme) in the Maho-Rambe area, and Wilgamuwa wewa (32 ha minor irrigation scheme) and Kandegedera (a rainfed paddy tract of 41 ha) in Maharachichimulla area. Finally, seven to eight non-SRI farmers were selected from each village. The structured questionnaire survey was implemented in January and February 2003 by the research team with the help of trained enumerators.

6

2.2

Agronomic and soil analyses

Following the completion of the socioeconomic survey, agronomic and soils studies were done during Yala4 2003 and Maha5 2003/2004 seasons. During the Yala 2003 season, agronomic and soils analyses were done in both Krunegala and Rathnapura districts, while during maha season the study was confined to Krunegala area. For the yala study, 10 SRI and 10 non SRI farmers were selected from survey respondents in each area. SRI farmers had practiced SRI in the same field for at least two seasons prior to sampling. Some of the SRI farmers have been practicing SRI since the year 2000. Soil samples were taken and analyzed pre-planting (see 2.2.1) and a limited set of yield samples were taken to validate the survey responses related to yield and post harvest quality. During maha season ten farmers who practiced both SRI and conventional systems of rice cultivation in the previous yala season (2003) were selected and supplied with weeders and seed paddy (variety BG 2/379) to be used in both SRI and conventionally managed fields. The specific tasks implemented under this research sub component are described as follows. 2.2.1

Soil sample collection and analyses

A composite surface soil sample (0-15cm) consisting of 15 random cores was collected from each SRI and conventional field just after first plowing. Samples were air dried and sieved through 2mm sieve prior to analysis. The collected samples were analyzed for soil pH, electrical conductivity (EC), inorganic nitrogen (N), exchangeable potassium (K), available phosphorus (P) and soil texture. The soil pH was measured using a pH meter with a glass electrode. A soil suspension of 1:2.5 soil and distilled water was used for the measurement as described by Mclean (1982). Electrical conductivity was measured using a conductivity meter with a saturation paste as described by Rhodes (1982). N, K, and P were determined on five gram subsamples of air dried soil according to the following standard methods: exchangeable (available) N by steam distillation with MgO and Devarda alloy (Keeney and Nelson, 1982, Page et al., 1982); Exchangeable K by ammonium acetate extraction and flame photometer as described by Knudesen et al. (1962): and available P by the Olsen’s extraction method (Olsen and Sommers, 1982). Soil texture was measured using the pipette method (Gee and Bauder, 1986), on 40 gram samples of air-dried soil. The pipetting times were calculated for silt and clay. The clay and silt contents were calculated and sand estimated as 100 – (sand+clay). Using these values the textural class was determined by textural triangle.

4 5

Yala is a Sinhala word for the short rainy season in Sri Lanka Maha is a Sinhala word for the long rainy season in Sri Lanka

7

2.2.2 Agronomic monitoring and measurement Yield and post harvest quality data were collected in the yala 2003 season. In the maha season, no yield and post harvest measurements were done due to crop failure resulting from the extreme drought. During yala 2003, yield data was obtained from a total of 12 farmers’ fields (Seven SRI fields and five conventional fields). In each field three one-meter square plots were randomly marked and harvested. From the harvested samples, 10 panicles where randomly selected to assess grain fill and number in SRI and conventional systems. The samples were taken to University of Peradenya, department of soil sciences, Sri Lanka for analyses. Tillering potential of differently grown rice plants is often cited as a major difference between SRI and conventionally grown rice (Nissanka and Bandara, 2004; Satyanarayana, 2004). To test this for farmers’ fields, tillering was evaluated in the Kurunegala maha 2003/2004 experiments using counts of number of tillers per hill and per square meter at flowering stage. For the number of tillers per hill, average tiller counts of a randomly selected twelve hills from each of the SRI and conventional fields were taken. To determine tiller counts per square meter, five one square meter area was randomly selected (each square meter separately and independently) from SRI and conventional fields at the flowering stage. During maha season of the year 2003/2004, the study locations in Krunegala faced about two and half months of severe drought, allowing us to do a preliminary assessment of differences in drought tolerance between rice grown under SRI and conventional systems under natural conditions. Drought tolerance of tillers in both SRI and conventional methods was evaluated by counting the survived number of panicle bearing tillers per square meter, on date, after experiencing extreme drought stress for 75 days. The counts were taken from a random sample of five one square meter areas at harvesting stage. All data were subjected to Univariate Analysis of Variance using SPSS’s General Linear Model (SPSS, 1999).

3 3.1

Results and discussions Farming practices, water management and agronomic inputs in SRI and conventional fields

Farmers’ agronomic practices and irrigation water management differed by production system and seasons as shown in table 2. There were significant differences between the two systems of rice production in planting method, soil fertility management, weed control practices and level of irrigation water application with consequent implication for resources conservation, environment and human health. These differences were measurable despite significant variation in practices that was found across

8

different farmers’ fields. On average a SRI farmer transplants one 8-day-old rice seedling per hill, on square grid of about 24 x 24 cm. While conventional farmers who did transplant, (most conventional fields were broadcast planted) planted on average four 18-day-old rice seedlings in a clump per hill on a square grid of 15x15 cm. Due to the difference in planting method, there was a very large difference in seeding rate, with SRI farmers using only approximately 13% of the seed as that used by conventional farmers. Table 2. Input utilization by production system SRI Practices/inputs

Conventional

Yala Mean

Maha SD

Mean

Ya la

SD

Mean

F value & significanceb

Maha

SD

Mean

SD

Number of seedlings/hill

1.1

0.3

1.4

0.7

4.1

1.3

3.8

1.1

275.93***

Seedling age in days

7.6

1.9

8.4

2.7

18.0

3.5

18.5

4.1

349.07***

Plant spacing in cm

24.1

3.3

24.0

3.2

15.0

3.9

15.1

3.3

174.70***

Row spacing in cm

24.7

3.1

24.4

2.9

15.2

4.1

15.7

3.8

179.99***

Seed rate (kg/ha)

14.0

10.0

17.0

12.0

117.0

38.0

114.0

38.0

369.7***

Herbicide (l/ha)

0.2

0.9

0.1

0.3

2.3

3.3

2.1

3.1

27.8***

Frequency of irrigation

24

17.9

22

17.2

32

25.9

29

25.8

5.3**

Nutrients from organic sources (kg/ha) : 58.9 Nitrogen 64.5

38.8

92.0

10.7

49.0

8.8

22.6

33.949***

Phosphorous

22.2

24.7

14.5

26.8

3.2

11.0

3.5

12.3

49.655***

Potassium

38.6

46.7

21.0

46.7

6.5

21.0

7.3

18.0

40.737***

a

Nutrients from inorganic sources (kg/ha): Nitrogen

66.2

63.4

57.9

55.1

109.1

55.1

104.5

50.5

35.688***

Phosphorus

10.5

19.4

8.2

13.7

25.0

55.1

20.0

14.6

6.9***

Potassium

25.5

41.8

18.3

25.8

37.7

25.0

36.9

24.0

17.351***

Total nutrient inputs (kg/ha): Nitrogen

125.1

127.8

96.6

147.1

119.8

104.1

113.3

73.1

0.233

Phosphorous

32.7

44.1

22.7

40.4

28.2

66.1

23.5

26.9

0.094

Potassium

64.1

88.7

39.3

72.4

44.2

46.0

44.3

42.0

1.569

a The main sources of organic fertilizers include cow dung, tree leaves, straw, poultry manure, compost and rice bran. b. The values are for the mean differences between SRI and the conventional system of paddy cultivation with respect to the parameters indicated in column one **, ***means the that the mean differences are significant at 5% and 1% level of significance

9

Significant differences were also observed in soil fertility management between the two systems, SRI farmers used lower rates of inorganic and higher rates of organic fertilizer per unit area, than conventional farmers. The farmers applied relatively more fertilizer during yala season irrespective of the type of production system. Overall, there is no appreciable difference in the total fertilizer (i.e., considering both organic and inorganic fertilizers) applied to SRI and conventional paddy fields. The major difference lies in the source: SRI fields receive more organic fertilizers and less inorganic fertilizer than conventional fields. The weed control method was another important aspect differentiating the two production systems. SRI farmers rarely used herbicides; instead they made use of a mechanical weeder or hand weeding, which on average required 311 hours of human labor per ha. The corresponding figure for conventional paddy fields was on average only 12.3 hours. The conventional farmers relied heavily on herbicides and they flooded their paddy fields with water as a means of weed control. On average the SRI farmers reduced herbicide use by 95%. Water management also differed significantly, and on average SRI fields received about 24 percent less irrigations. Overall external inputs to SRI fields in all categories, water, inorganic fertilizers, and pesticides were much lower than to conventional fields, demonstrating that in practice, and not just as recommended, the SRI system is a low external input system.

Micro-economics of the two systems of rice production was assessed using the enterprise budgeting technique. The results indicate that when costing the labor input at the prevailing farm wage rate, the net benefits for SRI increased by about 90-117%, while the per kilogram cost of production declined by 1727% (Namara et al., 2003). Studies from India and Cambodia show almost similar results. In Cambodia 74.2% increases in net benefits was reported (Anthofer, 2004) and in India 69.5% increases in net benefits was recorded (Sinha and Talati, 2004).

Most of the SRI farmers were only putting a portion of their land under SRI. The proportion of paddy area allotted to SRI ranges from about 39 percent in Kalthota to about 61% in Kurunegala. These farmers are operating within the context of country wide practices which in general have been moving in opposite directions from SRI recommendations. Island wise there has been remarkable changes in the paddy production process over the last two decades. (See figures 2, 3 and 4). Beginning in the late 1980s, the proportion of farmers broadcast sowing rice has shown an increasing trend, while the proportion of farmers using other planting methods such as transplanting and row seeding has been declining (see figure 2). 10

Percentage of pady farmers

Fig. 2 Island wise changes in rice planting methods 100.0 90.0 80.0

Maha: Broadcast Sowing

70.0 60.0 50.0 40.0 30.0

Maha: Transplanting & Row Seeding Yala: Broadcast Sowing Yala: Transplanting & Row Seeding

20.0 10.0 0.0 1980 1985 1989 1994 1997 2000 2003 Year

Source: Sri Lanka Department of Census and Statistics (various issues) Similarly, there is a notable change in weed control methods (figure 3). Beginning from the early 1980s, the proportion of rice farmers using manual hand weeding and flooding (impounding paddy fields with water) as a weed control strategy has sharply declined, while the proportion of farmers using herbicides has risen. The underlying reason for these changes in agronomic practices is the increase in the opportunity cost of labor (Weerahewa, 2004). Fig. 3 Island wise changes in weeding practices 90.0

Percentage of farmers

80.0 70.0

Maha: Hand Weeding

60.0

Maha: Herbicides

50.0

Maha: Water/Flooding

40.0

Yala: Hand Weeding

30.0

Yala: Herbicide

20.0

Yala: Water/Flooding

10.0 0.0 1980 1985 1989 1994 1997 2000 2003 Year

Source: Sri Lanka Department of Census and Statistics (various issues)

11

A slight change in the soil fertility management over the last two decades is also is notable (see figure 4). Since the late 1990s, the proportion of farmers using chemical fertilizer has shown a slight decreasing trend, while the proportion of farmers using organic fertilizer has slightly risen. The main reason for this is the rise in chemical fertilizer prices due to the economic liberalization policies that removed or substantially reduced input subsidies (FAO, 2000).

Percentage of the farmers

Fig. 4 Island wise changes in soil fertility management 100.0 90.0 80.0 70.0 60.0 50.0 40.0 30.0

Maha: Chemical Fertilizer Maha: Organic Fertilizer Yala: Chemical Fertilizer Yala: Organic Fertilizer

20.0 10.0 0.0 1994

1997

2000

2003

Year

Source: Sri Lanka Department of Census and Statistics (various issues) 3.2 Yield and yield component comparisons One of the attributes of SRI highly contested among researchers and others is its yield advantage over the conventional method of rice production (Stoop and Kassam, 2005). Some studies report a yield improvement of 19 to 270% due to SRI with absolute yield levels as high as 15 to 20 t/ha (McHugh et al., 2002 and Bonlieu, 1999; Stoop et al., 2002). On the other hand, some claim that SRI has no major role in improving rice production (Sinclair and Cassman, 2004; Sheehy et al., 2004). Our yield estimate from both socioeconomic survey (farmers’ report) and the crop cutting exercise indicates that, on average, SRI fields recorded higher yields than the conventional fields. The increase was statistically significant (see table 3), but not in the range of the very high potential yields reported by others (McHugh et al., 2002 and Bonlieu, 1999). Table 3. Yield, yield components, post harvest qualities and drought tolerance indicators SRI

Agronomic traits Mean

Conventional SD

Mean

SD

F value & significance

12

Yield from farmers’ reports a (kg/ha)

5524

2763

3836

1922

4.74**

Yield from crop cutting (kg/ha)

6365

2170

4778

2223

3.567*

Weight of seeds from 10 panicles in g

21.9

4.9

17.4

7.4

4.053*

Weight of filled seeds from 10 panicles in g

19.8

5.8

15.8

7.4

2.765

Weight of empty seeds from 10 panicles

1.7

0.9

1.6

0.4

0.025

Mean number of grains per panicle

115.6

19.7

87.4

26.0

11.261***

Percentage of grain filling

89.5

11.2

88.7

5.3

0.039

Thousand seed weight in g

19.2

3.8

19.2

3.2

0.000

Grain straw ratio

1.2

0.3

0.9

0.3

4.613**

22.9

2.5

9.1

1.0

260.48***

368.8

38.9

324.4

37.5

6.743**

b

Tillering count per hill at flowering stage 2

Tillering count per m at flowering stage

a Yield estimate from socioeconomic survey based on three years average (2000 to 2002) for both Maha and Yala seasons; b Yield estimate from crop cutting exercise (Yala 2003 season) based on 12 on-farm trial sites *,**,*** means that the mean differences between SRI and conventional system of rice cultivation with respect to the variables indicated in column one are statistically significant at 10%,5% and 1% level of probability

Comparison of the yield components (as an indicator of yield potential) also supports the observed significant yield improvement. The tillering count per hill for SRI fields is more than double than that of conventional rice plants. And tillering count per m2 for SRI is also significantly higher than that of conventional fields. Thus, apparently, enhanced tillering potential more than compensates for the wider spacing and lower rice plant density in the SRI fields. Increased tillering has often been reported for SRI (Nissanka and Bandara, 2004; Stoop et al., 2002) and is attributed to the younger age of seedlings at the time of transplanting that allows greater development of individual plants, and increased aeration of the soil due to intermittent irrigation and mechanical weeding. The grain straw ratio and the number of grains per panicle were also significantly higher for SRI. There were no significant changes in the grain quality parameters measured, percentage of grain fill and thousand seed weight. This is in contrast to the perception of farmers who consistently report that rice grown by the SRI method is of higher quality and findings from experimental studies (Namara et al., 2003; ANGRAU, 2004). It must however be noted that the measured parameters represent a very limited set of quality factors.

Also relevant from water resources management and production risk mitigation point of view is the effect of SRI on the drought tolerance of rice plants. In the very harsh drought condition of 2003/2004 maha season in Sri Lanka, SRI fields recorded significantly higher survival of panicle bearing per m2 tillers

13

(80% survival) at harvesting stage as compared to conventional rice fields (70% survival) (Table 4). Apparently the SRI package of practices increased the vigor of rice plants against the drought compared to the conventional system of rice cultivation. The mechanism by which this vigor was attained may lie in the effect of organic matter additions which can help increase soil water holding capacity, and increased root growth that has been reported for SRI (ANGRAU 2004). Although this evidence was sparse, it does indicate that drought tolerance is an area that warrants further investigation. Table 4. Tiller counts at flowering and under severe moisture stress in SRI and conventional fields during the 2003/2004 maha season. Tiller counts

SRI

Conventional

F-Value and significance

Mean

SD

Mean

SD

368.8

38.9

324.4

37.5

6.743**

Number of panicle bearing tillers per m at harvesting stage after experiencing extreme drought stress for 75 days

294.3

28.32

225.9

26.84

30.728***

% of panicle bearing tillers surviving

79.8%

2

Tillering count per m at flowering 2

69.6%

**, ***means that the mean differences are significant at 5% and 1% level of probability 3.3 Effects on some soil parameters Soil fertility management is one area of significant difference between SRI and conventional rice production. SRI farmers used significantly more organic matter inputs than conventional farmers, who favoured inorganic fertilizers as their primary input. Texture did not vary among sites indicating that soil type was similar at all sites (Table 5). Soil pH was significantly higher in SRI fields, and EC was significantly lower. Differences may be attributed to input type, and introduction of salts with inorganic fertilizers. Regardless, differences were small, and soil pH and EC level in all fields were within optimal and non-limiting levels for growth of rice. Soil organic carbon levels did not change. This is not surprising considering the rapid rate of organic matter cycling in this tropical climate.

Of the plant

macronutrients there was no significant difference in inorganic N, there were significantly higher levels of available phosphorus, and there was a trend (significant at the 10% level) toward higher potassium in SRI fields as compared to conventional fields. These results are consistent with others (Clark, et al. 1998) looking at changes in soil chemical properties in organic and conventional system comparisons. Increases in phosphorus and potassium, represent a small improvement in the fertility of the soils after 2 seasons of SRI management. Table 5. The influence of the SRI practice on soil physical and chemical properties Soil parameters

SRI

Conventional 14

Mean

SD

N

Mean

SD

N

Organic Carbon (%)

1.36

0.49

12

1.34

0.30

12

0.018

Exchangeable Potassium (ppm)

75.3

40.9

28

61.6

25.2

28

2.264*

Available Phosphorus (ppm)

14.6

1.7

27

10.1

1.6

28 104.825***

Available Nitrogen (Nmg/100g of soil)

5.5

1.8

28

5.7

1.9

28

0.189

Soil pH (1: 2.5 soil: water)

6.1

0.3

18

5.7

0.6

18

5.953**

0.039 18

9.738***

0.021 18

0.068

F value & significance

EC (m mhos/cm)

0.035

Sand content (%)

71.9

8.8

18

72.1

7.6

18

0.002

Silt content (%)

12.0

5.5

18

13.1

4.5

18

0.399

Clay content (%)

15.8

4.3

18

13.7

4.9

18

1.994

*,**, ***Means that the mean differences are significant at 10%, 5% and 1% level of probability

4. Concluding remarks SRI as adopted by farmers in Sri Lanka included changes in practices in all four elements of SRI, planting method, soil fertility management, weed control, and water (irrigation) management. On SRI farms, irrigations were reduced by 24%, seeding rates were reduced by 85%, and plant spacing was increased by 60%. Total inputs of nitrogen, phosphorus, and potassium were similar across SRI and conventional plots, but the source of nutrients was different. SRI farmers reduced herbicide usage by 95%. Yields were variable, but significantly higher on SRI farms, soil available potassium and phosphorus were increased, and SRI plants exhibited better tolerance to low moisture stress. SRI as practiced, and not just as recommended, was found to provide many of the potential benefits as claimed (refs), including lower requirements for external inputs without negative impacts on yield. The SRI resulted in more effective tillers per unit area and saved about 0.1 ton of seed per hectare, the latter having potentially significant household food security benefits. It was observed that in the advent of severe drought SRI fields exhibited more tolerance to low moisture stress than the conventional fields with consequent implication for both rain-fed and irrigated rice production systems.

For an individual farmer or household cultivating rice in rainfed or water stressed systems the most important benefits of SRI are likely to be reduced input costs that reduce risk due to paddy price volatility, and yield loss, stabilization of yields given uncertain water availability, improved soil fertility, and reduced pesticide use leading to better health. Poisoning resulting from mishandling of pesticides is a major health problem in Sri Lanka and other Asian countries (Pingali et al., 1994). One third of all deaths 15

in hospitals in southern Sri Lanka are the result of acute pesticide poisoning. (van der Hoek and Konradsen, 2005). Benefits of reduced pesticide and fertilizer inputs are also felt at a larger scale, when agricultural pollution and energy demands are reduced. However, potential larger scale benefits of reduced water use by individual farmers in irrigation systems have been questioned. To achieve water savings at the larger scale would require that individual management be aggregated up to the system level, allowing alternative allocation of water resources.

However, in many cases, there has already been a

reallocation of water, or an under allocation to the irrigation systems, so the challenge is not so much to ‘save’ water for other uses, but to produce rice with the sub optimal quantities of water that are available.

The greatest disincentive for SRI can be found by looking at national trends in rice management practices. The primary driving force for changes including shifts towards broadcast seeding and increased use of herbicides for weed control is the increasing opportunity cost of labor. SRI recommendations are opposite to most national trends, and widespread adoption is not likely beyond the identified target group of very resource poor farmers (Namara et al. 2003), or where the larger scale benefits are great enough to motivate promotion and extension efforts. One such example is the Kalthota irrigation system, where the Ceylon Electricity Board took steps in promoting SRI among farmers in the irrigation command, in order to mitigate negative consequences of diverting water for hydropower production. Another important trend internationally that may affect promotion of integrated low input systems such as SRI are concerns for consumer safety in light of excessive use of pesticides in developing countries. Rice has received considerable attention in this regard just recently.

The south Asia region is currently experiencing surplus grain production, including paddy rice. However, on a local level increasing and/or stabilizing paddy production is a critical livelihood concern for many poor farmers (Bouman et al.2007). In the longer term, increasing populations will result in increasing demands for rice and other staple crops. Increasing productivity is a challenge, due to a lack of available land and water resources to expand production, and the unreliability of water, particularly in rainfed paddy rice systems. To this end there are a whole range of inter-related rice production systems currently being promoted or tried (Barker et al., 2001). SRI is one such system that is attractive to poor farmers (Namara et al. 2004). Further work to characterize drought tolerance and water conserving benefits of the system is warranted to better define the conditions under which this package of practices is most useful.

16

5. References ANGRAU (Acharya N.G. Ranga Agricultural University)(2004): Manual on “System of Rice intensification” SRI: a revolutionary method of rice cultivation. 1st edition. Rajendranagar, Hyderabad-500 030, India. ANTHOFER, J. (2004): The potential of the System of Rice Intensification (SRI) for poverty Reduction in Cambodia. Deutscher Tropentag 2004 October 5-7, Berlin. BARKER R, LOEVE R, LI YH, AND TUONG TP (EDS.) (2001):Water-saving irrigation for rice. Proceedings of an international workshop held in Wuhan, China, March 23-25. Colombo, Sri Lanka: International Water Management Institute. BONLIEU, F. (1999): Summary of findings from thesis (http://ciiad.cornell.edu/SRI/bonlieu.pdf) accessed: December 5, 2006.

research

on

SRI

BOUMAN, B., R. BARKER, E. HUMPHREYS, T.P. TUONG, G.ATLIN, J.BENNET, D.DAWE, K.DITTERT, A.DOBERMANN, T.FACON, N.FUJIMOTO, R.GUPTA, S.HAEFELE, Y.HOSEN, A.ISMAEL, D.JOHNSON, S.JOHNSON, S.KHAN, L.SHAN, I.MASIH, Y.MATSUNO, S.PANDEY, S. PENG, T.M.THIYAGARAJAN AND WASSMAN, R. 2007. Rice: Feeding the billions. In Molden, David (Ed.). Water for food, water for life: A Comprehensive Assessment of Water Management in Agriculture. London, UK: Earthscan; Colombo, Sri Lanka: IWMI. pp.515-549. CLARK, M.S., HORWATH, W.R., SHENNAN, C., and SCOW, K.M. (1998): Changes in soil chemical properties resulting from organic and low-input farming practices. Agronomy Journal 90, pp. 662671. DHANAPALA, M.P. (2000): Bridging the rice yield gap in Sri Lanka. In: Bridging the rice yield gap in the Asia-Pacific region, Papademetriou MK, Dent FJ, and Herath EM (Eds.). Regional office for Asia and the Pacific. FAO: Rap publication 2000/16. ELLIS, F. (1993): Peasant economics: Farm households and agrarian development. Cambridge university press. EVANS, J.R. (1989): Photosynthesis and nitrogen relationship in leaves of C3 plants. Oecologia 78:9-19. EVENSON, R.E., and GOLLIN, D. (2003): Assessing the impact of the Green Revolution, 1960 to 2000. Science 300: 758–762. FAO (1994): Land degradation in south Asia: Its severity, causes and effects upon the people. Food and Agriculture Organization of the United Nations, Rome, 1994. FAO (2000): Agriculture, trade and food security issues and options in the WTO negotiations from the perspective of developing countries,Vol. II, country case studies. Commodities and trade division, Food and Agriculture Organization of the United Nations, 2000. FAO (2002): World agriculture: towards 2015/2030. Summary report. Economic and Social Department, FAO, Rome. FOX, T.R., COMERFORD, N.B., MCFEE, W.W. (1990):Kinetics of phosphorus release from Spodosols: Effects of oxalate formate. Soil Sci.Soc. Am J. 54:1441-1447. GEE, G.W., AND BAUDER, J.W. (1986): Particle Size Analysis. Black CA (Ed.) Methods of soil analysis Part I. 2nd Edition. American Society of Agronomy, Madison, Wisconsin: 413-422.

17

HSAIO, T.C., and LAUCHLI, A. (1986): Role of potassium in plant water relationship. In: Advance in plant nutrition, Vol 2. Tinkler B, and Lauchli A (Eds.) Prager Publisher, New York:281-311 IFPRI (2002): Green Revolution: curse or blessing? International Food Policy Research Institute (IFPRI), Issue Brief No. 11. (www.ifpri.org/pubs/ib/ib11.pdf). IRRI (1997): Rice almanac. 2nd edition. Los Banos (Philippines): International Rice Research Institute, 181p. JENKINSON, D.S., LADD, J.N. (1981): Microbial biomass in soil: measurement and turnover. Soil biochemistry, Vol. 5. Dekker, New York: 415- 471. KNUDSEN, D., PETERSON, G.A., and PRATT, P.F. (1982): Lithium, Sodium, Potassium. Page AL (Ed.). Methods of Soil Analysis Part II. 2nd Edition. American Society of Agronomy, Madison, Wisconsin: 225-245. LIPTON, M., and LONHURST, R. (1989):New seeds and poor people. London: Unwin Hymen. MCHUGH, O.V., STEENHUIS, T.S., BARISON, J., FERNANDEZ, E.C.M., UPHOFF, N.T. (2002): Farmer implementation of alternate wet and dry and non-flooded irrigation practices in the system of rice intensification. In: Proceedings of the international workshop on water-wise rice production, Los Banos, Philippines, April 8-11:89-102. Mclean EO (1982) Soil pH and lime requirement. Page AL (Ed.). Methods of soil analysis Part II. 2nd Edition. American Society of Agronomy, Madison, Wisconsin, pp. 199-223. NAMARA, R.E., WELIGAMAGHE, P., BARKER, R. (2003): Prospects for adopting system of rice intensification in Sri Lanka: A socioeconomic assessment. Research Report 75. Colombo, Sri Lanka: International water Management Institute. NISSANKA, S.P., and BANDARA, T. (2004): Comparison of productivity of system of rice intensification and conventional rice farming systems in the dry zone region of Sri Lanka. Poster presentation in the 4th International Crop Science Congress. (www.regional.org.au/au/cs/2004/poster/1/2/117__nissankara.htm) OLSEN, S.R., and SOMMERS, L.E. (1982): Phosphorous. Page AL (Ed.) Methods of soil analysis Part II. 2nd Edition. American Society of Agronomy, Madison, Wisconsin;199-223 PAGE, A.L., MILLER, R.H., and KEENEY, D.R. (1982) Methods of Soil Analysis Part II. 2nd Edition. American Society of Agronomy, Madison, Wisconsin. PINGALI, P.L., MARQUEZ, C.B., PALIS, F.G.(1994):Pesticides and Philippine rice farmer health: Medical and economic analyses. American Journal of Agricultural Economics 76(3):587-592. RHOADES, J.D. (1982): Soluble salts. Page AL (Ed.) Methods of soil analysis Part II. 2nd Edition. American Society of Agronomy, Madison, Wisconsin:167-178. ROY,

I. (2003): Penury in plenty. In: Himal (www.himalmag.com/2003/July/analysis.htm).

South

Asian

magazine,

July

issue.

SATYANARAYANA, A. (2004): Rice, research and real life in the field. Nature 429, p.803. SHEEY, J.E., PENG, S., DOBERMANN, A., MITCHELL, P.L., FERRER, A., YANG, J., ZOU,Y., ZHONG, X., HAUNG, J. (2004): Fantastic yields in the system of rice intensification: Fact or fallacy? Field Crops Research 88:1–8. SINHA, S.K., and TALATI, J.(2004): Impact of System of Rice Intensification (SRI) on rice yields. Results from a new sample study in Purulia District, India. IWMI-TATA Water Policy Research Highlight 20

18

SICLAIR, T.R., CASSMAN, K.G. ( 2004): Agronomic UFOs? Field Crops Research 88: 9-10. SINGH, I. (1990): The Great Ascent: The rural poor in south Asia. Baltimore: Published for the World Bank, the John Hopkins University Press. SRI LANKA DEPARTMENT reports. Colombo.

OF

CENSUS

AND

STATISTICS (Various Issues): Seasonal paddy cultivation

SPSS (1999): Statistical Package for the Social Sciences. SPSS version 10. Chicago, Illions: SPSS inc. STOOP, W.A., KASSAM, A.H. (2005): The SRI controversy: a response. Field Crops Research 91: 357-360. STOOP, W.A., UPHOFF, N., and KASSAM, A. (2002):A review of agricultural research issues raised by the system of rice intensification (SRI) from Madagascar: opportunities for improving farming systems for resource-poor farmers. Agricultural systems 71: 249-274. TENNAKOON, M.U.A. (1986): Drought hazard and rural development: a study in perception of and adjustment to drought. Central Bank of Sri Lanka THIYAGARAJAN, T.M., VELU, V., and RAMASAMY, S. (2002): Effects of SRI practices on hybrid rice performance in Tamil Nadu. In: Proceedings of International Conference on Water Wise Rice Production, Los Banios Philippines, April 8-11:119-128. VANDANA, S. (1991):The violence of the green revolution: third world agriculture, Ecology and politicsLondon (u.a): Zed Books. UPHOFF, N. (2002): Systems of rice intensification (SRI) for enhancing the productivity of land, labour and water. Journal of Agric. Resource Management 1(1):43-49. VAN DER HOEK, W., and KONRADSEN, F. (2006): Analysis of 8000 hospital admissions for acute poisoning in a rural area of Sri Lanka. Clinical toxicology, Volume 44(3): 225-231. WARDLE, D.A., PARKINSON, D. (1990): Comparison of physiological technique for estimating the response of soil microbial biomass to soil moisture. Soil Biol. Biochem. 22: 817-823. WEERAHEWA, J. (2004): Impacts of trade liberalization and market reforms on the paddy/rice sector in Sri Lanka. MTID discussion paper No. 70. Markets,Trade and Institutions Division. IFPRI, 2003 K Street, N.W. Washington, D.C. 2006 USA.

19