Paddy and Water Environment https://doi.org/10.1007/s10333-018-0646-y
ARTICLE
Assessing the acceptance of the system of rice intensification among farmers in rainfed lowland rice region of Cambodia Yun Ho Lee1 · Kazuhiko Kobayashi1 Received: 28 September 2017 / Revised: 25 March 2018 / Accepted: 5 April 2018 © The International Society of Paddy and Water Environment Engineering and Springer Japan KK, part of Springer Nature 2018
Abstract We conducted interview survey with rainfed rice farmers at a commune in southern Cambodia to identify the determinants of their adoption or rejection of the system of rice intensification (SRI) for the wet season of 2014. SRI was first introduced to the commune in 2003 and spread among the farmers, but the number of farmers discontinuing the practice has increased since 2011. We classified the farmers into four categories: those who practice SRI in all their fields (OA), farmers who practice SRI in some of their fields (PA), farmers who had practiced SRI but discontinued it (DA), and farmers who have never practiced SRI (CR). Farmers in different categories of SRI adoption differed in their access to water sources. The majority of OA and PA farmers had supplementary water sources, which was, however, available to less than 50% of DA farmers and only 15% of CR farmers. Both PA and DA farmers mentioned water shortage as the reason for not practicing SRI in some (PA) or any (DA) of their fields, but they differed in the number of rice fields. All the PA farmers had more than one field, whereas a majority of DA farmers had only one field. Labor shortage and difficulty of planting in a regular grid pattern were also mentioned as the reason for not practicing SRI by the farmers, particularly by DA farmers. Water constraint and difficulties in transplanting were thus identified as two major determinants of SRI adoption/rejection at the study site. Keywords SRI · System of rice intensification · Cambodia · Water · Rainfed rice
Introduction The system of rice intensification (SRI) established by Fr. Henri de Laulanie in Madagascar in the early 1980s has primarily targeted increasing the rice harvest for smallholder farmers and using less water and seeds without relying on external inputs such as chemical fertilizers. Over the past 30 years, many studies comparing SRI and conventional rice cultivation practices have indeed reported yield increases for SRI in the range of 20–40% with 20–50% less water use and 50–70% lower seed costs (Ceesay et al. 2006; Sinha and Talati 2007; Menete et al. 2008; Stoop et al. 2009; Adusumilli and Bhagya Laxmi 2011; Styger et al. 2011; Ndiiri et al. 2013; Islam et al. 2014; Thakur et al. 2016). It has also been reported that SRI increased the economic return per hectare by 76% (Balamatti and Uphoff 2017). * Kazuhiko Kobayashi
[email protected] 1
Graduate School of Agricultural and Life Sciences, The University of Tokyo, 1‑1‑1 Yayoi Bunkyo‑ku, Tokyo 113‑8657, Japan
The advantages of SRI over conventional practices could explain its promotion by both non-governmental organizations (NGOs) and governmental institutions in many developing countries where rice is the major agricultural product (Kabir and Uphoff 2007). For example, SRI techniques were very quickly promoted and adopted by NGO and national government institutions in Cambodia (Sam and Ouch 2015). Cambodia is one of the countries where SRI has spread fast in rural areas. In 2000, only 28 farmers participated in the first SRI experimentation organized by the Cambodian Center for Study and Development in Agriculture (CEDAC) (Yang 2002). Since then, for the 13 years from 2000 to 2012, the area of fields with SRI adoption has increased from 1.6 ha to 101,719 ha (Sothy 2008; SRI-LMB Project 2013). SRI was included in the National Strategic Development Plan (NSDP) for 2006–2010 to raise productivity in the rice sector, as well as in the revised NSDP for 2009–2013 (MAFF 2015). Despite the wide acceptance of SRI, it is also known that the dissemination of SRI is often accompanied by partial adoption, discontinuance, and continued rejection among farmers in Cambodia (Ly et al. 2012), as well as
13
Vol.:(0123456789)
Paddy and Water Environment
other rice-growing countries (Moser and Barrett 2003, 2006; Senthilkumar et al. 2008; Noltze et al. 2012; Takahashi 2013; Tumusiime 2017). Therefore, the uptake of SRI among Cambodian farmers is not a uniform acceptance of a superior technology by all farmers across the wide spectrum of natural and socioeconomic environments. Indeed, Lee and Kobayashi (2017) found that the yield increase by SRI is dependent on availability of supplementary water particularly at the beginning of rice growth in rainfed fields of southern Cambodia. They suggested that the farmers have recognized the dependency of SRI-induced yield gain on water availability when practicing SRI in their fields and the lack of water affects the uptake of SRI. In this study, we investigate the acceptance/uptake of SRI by farmers under various environmental conditions including supplementary water availability at the same location as Lee and Kobayashi (2017) in a rainfed rice region of Cambodia. We first categorize the farmers according to their adoption of SRI, such as practicing SRI in all their fields or discontinuance of SRI after some years of adoption. We then compare the natural, agronomic, and economic environments of the farmers among the SRI adoption categories. We also analyze the farmers’ recognition of positive and negative aspects of SRI. We intend in this study to understand the determinants of the farmers’ decision on adoption (or rejection) of SRI and, thereby, to confirm the link between the farmers’ decision and the hydrological constraint as suggested by the study on individual fields (Lee and Kobayashi 2017). Better
understandings of farmers’ adoption of SRI will clarify the prospect of SRI in increasing rice production under the various natural and socioeconomic environments of rural Cambodia.
Site and methods of study Study site We conducted the study in the Popel commune (11°04′67″N, 104°40′79″E) of Tram Kak district of Takeo province in Cambodia. The district is located to the south of the Phnom Penh Municipality (Fig. 1). The rice production is dominated by single cropping rice in the wet season. According to the District Department of Agriculture (DDA), SRI was first introduced to Tram Kak district by CEDAC in 2000. In 2012, the district had 10,442 farm households’ using SRI. This large-scale adoption of SRI is the main reason for choosing this district for this study. Rainfall during the wet season (June to October) averaged 838 mm from 2003 to 2010 according to the Provincial Department Office of Water Resources and Meteorology. In 2011, however, it declined to 763 mm and has stayed below 750 mm from 2012 through to 2015 (Fig. 2b of Lee and Kobayashi 2017). Importantly, 2014 and 2015 were anomalously dry during the transplanting time (June and July) (Fig 3 of Lee and Kobayashi 2017).
Map of Takeo Province
Map of Cambodia
Popel Commune
Tram Kak District
Phnom Penh
0 Takeo
7
14 km
Provincial border
District border
Fig. 1 Location of the study site: Popel Commune, Tram Kak District, Takeo Province of Cambodia
13
Paddy and Water Environment
Interview surveys We conducted face-to-face interviews with 106 randomly chosen households in June and July 2015 and asked about their adoption record of SRI, e.g., year of starting (and if applicable, stopping) SRI, their views on SRI, and the features of their rice production such as harvested area and amount for the wet season of 2014. According to the record of SRI adoption, the farmers were classified into four categories: farmers who currently practice SRI in all their fields (OA: overall adoption = SRI only), farmers who currently practice SRI in some fields but not all their fields (PA: partial adoption = both SRI and non-SRI), farmers who once practiced SRI but do not practice it currently (DA: discontinued adoption), and farmers who have never practiced SRI (CR: continued rejection = non-SRI only). It must be noted that, in this categorization, the distinction between practicing SRI and otherwise is far from clear-cut. As noted by Lee and Kobayashi (2017), the rainfed fields did not allow the farmers to perform all the SRI components rendering their actual practice as a hybrid between SRI and conventional practices. No farmer fully followed the SRI principle in water management, whereas most SRI adopters followed the principles of transplanting seedlings and application of organic matter. Indeed, they considered that they are practicing SRI when they transplant one or two seedlings per hill in a regular grid pattern with a wider spacing at an earlier stage than they do conventionally. In conventional practice, by comparison, they transplanted from three to five seedlings per hill (aged 30–45 days) without adhering to the grid pattern spacing between hills. As noted by Lee and Kobayashi Table 1 Rates of adoption and rejection of SRI and rice harvest in 2014 for different categories of SRI adoption Number of interviewed HH Year of starting SRI 2000–2005 2006–2010 2011–2014 Year of discontinuing SRI 2000–2005 2006–2010 2011–2014 Rice harvest area (ha) (P = 0.269)* Rice harvest amount (t) (P = 0.011)* Rice yield (t ha−1) (P = 0.004)*
(2017), the definition of a field in the interview often actually represented a group of fields being sequentially located and managed similarly by the same farmer. To augment the first survey results, a second survey was conducted in August 2015. Households were randomly selected from each of the four categories of SRI adoption and presented a questionnaire which covered household characteristics, such as household size, number of family members working in factories, working factory period, and major sources of income. Of the 53 households involved in the second survey, we chose 47 households for the analysis excluding those with exceptionally large income from jobs like taxi driving. The interview survey data were analyzed by using JMP software (SAS Institute, Cary, USA). Statistical significance was reported by the probability of the type I error as P values, and an effect with P value less than or equal to 0.05 was described as ‘significant.’
Results Categorization of farmers according to SRI adoption and comparison between the SRI adoption categories of rice harvest Of the 106 households interviewed, 12 (11%) was OA, 13 (12%) was PA, 29 (27%) was DA, and 52 (49%) was CR farmers (Table 1). It must be noted that the adoption or rejection of SRI herein followed the farmers’ definition, where practicing SRI meant that they transplanted one or
SRI currently practiced
SRI not practiced currently
OA
PA
DA
CR
12 (11%)
13 (12%)
29 (28%)
52 (49%)
2 (17%) 3 (25%) 7 (58%)
6 (46%) 2 (15%) 5 (39%)
3 (10%) 11 (38%) 15 (52%)
0 0 0
0 0 0 0.81 a 2.94 a 4.12 a
0 0 0 0.78 a 2.53 ab 3.23 ab
0 (0) 1 (3%) 28 (97%) 0.70 a 2.03 ab 3.07 b
0 0 0 0.56 a 1.50 b 2.76 b
OA overall adoption, PA partial adoption, DA discontinued adoption, CR continued rejection *Statistical significance of the difference between SRI adoption types is shown in P value. Means sharing a common alphabetical letter are not significantly different (P = 0.05) according to the Tukey–Kramer HSD multiple comparison test
13
Paddy and Water Environment
100 80 Frequency (%)
two seedlings per hill in a regular grid pattern with wider spacing at an earlier stage than the conventional practice. As noted earlier, the SRI adopters did not fully follow the SRI principle with regard to water management. Most, if not all, farmers of the CR category had heard about SRI, although none of them has adopted it. While the rice harvest area did not differ between the SRI adoption categories, harvested amount and yield per unit area differed (Table 1). The mean rice yield for OA households was significantly higher than those of DA and CR households with that of PA being in the middle (Table 1).
60 40 20 0
Temporal trajectory of SRI adoption and discontinuation among the farmers In the Popel commune, SRI was first introduced by CEDAC in 2003. The number of SRI farmers increased from two households in 2003 to 42 households 9 years later, and it increased to 54 households by 2014 (Fig. 2). Discontinuance began in 2009, but the largest discontinuance occurred in the 2012–2013 period. Most farmers stopped SRI after practicing it for 1–3 years.
Objective and subjective determinants of SRI adoption, rejection, and discontinuation The increase in tiller number was mentioned by most farmers as the main reason to adopt SRI (Fig. 3). Higher yields and lower seed use followed being mentioned by about a half of the farmers, whereas labor saving was mentioned by only a small fraction of them (Fig. 3). On average, a household planted 2.7 fields in the study area and the area of planted rice fields was only about 0.43 ha per household. When a household had more than
High High tillering tillering
High yield SeedSeed saving Labor saving High Labor yield saving saving
Fig. 3 Reasons given by farmers of OA (overall adoption) and PA (partial adoption) categories (n = 25) for their adopting SRI
one field, they were more likely to have supplementary water than those who has only one field (Fig. 4; P = 0.009). The SRI adoption category was clearly related to the number of fields per household (P = 0.001) with the PA category being limited to the households having more than one field, while c. 70% of the DA farmers had only one field (Fig. 5a). The SRI adoption category was also clearly related to the availability of supplementary water (P < 0.001). The majority of OA and PA farmers had supplementary water, which was available to less than a half of the DA farmers and only 15% of the CR farmers (Fig. 5b). In contrast to the reasons for adopting SRI, those for not adopting SRI were more diverse (Fig. 6). By far the
% 100 Cumulative number of SRI adopters
50
40
Cumulative number of SRI discontinuers
30
20
Supplementary water
Number of adopters/ discontinuers
60
75
Not available
50 25
Available
0 =1
10
>=2
Number of fields 0
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Year
Fig. 2 Cumulative numbers of SRI adopters (filled circles connected with solid lines) and discontinuers (unfilled circles connected with dotted lines)
13
Fig. 4 Comparison between households with respect to number of fields and availability of supplementary water. The proportions on the horizontal axis represent the number of households for each category of number of fields, and those on the vertical axis represent the number of households for each category of supplementary water availability
Paddy and Water Environment
Number of fields
% 100 %
a
75
>= 2
50 =1
25 0
Supplementary water
100
b
75
Not available
50 25 Available
0 OA
PA
CR DA SRI adoption
Fig. 5 Comparison between SRI adoption categories of a number of fields per household and b availability of supplementary water. OA overall adoption, PA partial adoption, DA discontinued adoption, CR continued rejection. The proportions on the horizontal axis represent the number of households for each category of SRI adoption, and those on the vertical axis represent a number of fields per household and b the number of households for each category of supplementary water availability
most mentioned reason against SRI was the lack of labor for transplanting. The difficulty to transplant seedlings at a regular grid pattern using a rope was the next commonly mentioned reason. As noted later, in conventional practice, rice seedlings are planted at a pace and spacing determined by the individual planters. In SRI practice at the study site, by comparison, transplanting rope is used to attain the regular spacing between hills, but it forces the planters
Household characteristics and sources of incomes for household Analyzing the results of the second household survey, we found no statistical difference between the SRI adoption categories in the household characteristics or the household income (Table 2). When, however, we compared the households having incomes from rice and factory jobs with those having income from rice only, the former had higher income (3808 USD/year) than that of the latter (2857 USD/year) with a statistical significance (P = 0.016) (data not shown).
50 40
Frequency (%)
Fig. 6 Reasons given by farmers of PA (partial adoption), DA (discontinued adoption), and CR (continued rejection) categories (n = 94) for not adopting SRI
to proceed at the pace of the slowest planter, which they often feel difficult to follow. Both of these two reasons are related to the difficulties at transplanting. The other reasons mentioned by about 20% of the farmers include lack of experience, difficulty in leveling the land, and lack of water (Fig. 6). The subjective reasons for not adopting SRI differed between the SRI adoption categories. PA and DA farmers mentioned lack of water as the reason for not adopting SRI significantly more than CR farmers (Fig. 7a; P = 0.0011). In contrast, lack of experience was mentioned only by CR farmers (Fig. 7b; P < 0.001), and it was indeed the most often mentioned reason for them not to adopt SRI (data not shown). PA and DA farmers were thus similar to each other in mentioning the lack of water as the reason for not practicing SRI in some (PA) or any (DA) of their fields. DA farmers, however, diverged from PA farmers in that they mentioned the difficulties of SRI at transplanting: use of transplanting rope and labor shortage. Raising the issue of using the rope as an obstacle against practicing SRI differed significantly among the SRI adoption categories (P = 0.028): no PA farmers mentioned it, whereas one third (DA farmers) or one fourth (CR farmers) did so (Fig. 8a). Lack of labor at transplanting was mentioned across the SRI adoption categories, but more often by DA farmers compared to the other farmers (Fig. 8b; P = 0.055).
30 20 10 0
Lack of labor Hard to use rope No experience Hard to leveling Lack of water Hard to Hardship Lack of No exLack of for transplant
planting labor
use rope
perience
in leveling
water
Lack of rice Far from home Shortage of Too far stock for rice straw from livestock/ forcompost livestock home and compost
Problem in Problem topography
in topography
Others Others
13
Paddy and Water Environment
%
%
a
75
Not mentioned
50
100
No water
25
Planting with rope
Water shortage
100
b
50 25 Hard to use rope
75
100
No experience
50 Not mentioned 25 DA
CR
SRI adoption Fig. 7 Comparison between SRI adoption categories of the reason for farmers not adopting SRI to be a water shortage and b lack of experience. PA partial adoption, DA discontinued adoption, CR continued rejection. The proportions on the horizontal axis represent the number of households for each category of SRI adoption, and those on the vertical axis represent the number of households mentioning a water shortage or b lack of experience
Discussion Temporal change of SRI adoption The temporal change in SRI adoption exhibited two periods of rapid increase (Fig. 2). According to our observations, those who adopted SRI during the first period of increase from 2003 onwards were mostly the village office staff, village development committee members, and some of the farmers who learned the SRI technique from the NGOs. The second increase in SRI adoption since 2010 (Fig. 2) was due to the local NGO’s campaign, which encouraged the farmers to practice organic farming. However, discontinuation of SRI, which began as early as 2009, started to increase in 2011 and reached 30% in 2014 (Fig. 2). The increase in SRI discontinuation coincided with the declining trend of wet season rainfall since 2011 (Lee and Kobayashi 2017). Discontinuation of SRI has also been reported in other countries (Moser and Barrett 2003; Takahashi 2013). In the case of Madagascar (Moser and Barrett 2003), however, the occurrence of discontinuation of SRI corresponded to the temporary disruption of the extension services.
No labor for transplanting
50 Not mentioned 25 0
PA
b
75 Labor
Experience
100
13
Not mentioned
0
0
0
a
75
PA
DA
CR
SRI adoption
Fig. 8 Comparison between SRI adoption categories of the reason for farmers not practicing SRI related to difficulties in a the usage of rope and b labor requirement for transplanting. PA partial adoption, DA discontinued adoption, CR continued rejection. The proportions on the horizontal axis represent the number of households for each category of SRI adoption, and those on the vertical axis represent the number of households mentioning a hardship in planting with rope or b lack of labor
Environmental determinants of SRI adoption, discontinuation, and rejection It is evident that availability of supplementary water is among the major determinants of the farmers’ decision on SRI adoption. A majority of OA farmers had an available water supply in their fields, whereas most of the CR farmers lacked it (Fig. 5b). All PA farmers had two or more fields (Fig. 5a), and most of the farmers had supplementary water (Fig. 5b), while mentioning lack of water as a reason for not adopting SRI in some of their fields (Fig. 7a). A majority of DA farmers, in contrast, had only one field (Fig. 5a), and only less than a half of them had supplementary water in their field (Fig. 5b). Some of the DA farmers did mention the lack of water as a reason for not adopting SRI in their fields (Fig. 7a). The constraint of water shortages on SRI adoption was also evident in our own observations during the field visits. For instance, there was a severe drought from June through to August in 2014, when the farmers had water shortage for transplanting in nearly all their fields. Many farmers had to reduce their rice planting area or to plant just in a small area near their water sources, if any. The
Paddy and Water Environment Table 2 Household (HH) characteristics and sources of income for household from the second interview survey for different categories of SRI adoption
Number of interviewed HH HH size [female] HH members working for factory [female] Number of HH working for factory Work periods for factory (months) HH income per year (USD)
SRI currently practiced
SRI not practiced currently
OA
PA
DA
CR
5 (9%) 5.8 [2.8] 0.8 [0.8] 2 17 3064
8 (15%) 4.8 [1.8] 0.8 [0.4] 3 12 3550
15 (34%) 5.5 [2.9] 1.2 [1.1] 9 20 3623
19 (42%) 4.6 [2.2] 0.8 [0.6] 10 15 3108
P value
– 0.384 [0.125] 0.698 [0.337] 0.722 0.915 0.683
OA overall adoption, PA partial adoption, DA discontinued adoption, CR continued rejection
low rainfall early in the season also delayed the sowing and transplanting from July to August or September. We observed in early September that seedlings planted in the SRI fields in early August exhibited drought stress symptoms like leaf yellowing, but that no seedlings planted later in non-SRI fields developed such symptoms (Lee and Kobayashi 2017). Water shortage is thus a persistent trouble for cropping in rainfed lowland areas particularly under SRI management. For 2014, we also found that the increase in rice yield due to SRI was nonsignificant where no supplementary water was available (Lee and Kobayashi 2017). Other studies on SRI in Cambodia have also reported that the farmers found it hard to sustain water in the fields if they do not have ponds or streams in addition to rainfall (Tech 2004; Sam and Ouch 2015). Ly et al. (2012) have also noted limited water availability as a major constraint on SRI farmers’ agronomic practices in Cambodia. The adopters of SRI could not adhere to some of the SRI principles due to the water supply constraint, which is consistent to our findings. Ly et al. (2012) did not, however, mention the water constraint as a determinant of SRI adoption. This inconsistency is remarkable, since one of their study sites was in the same district (Tram Kak) as ours, but in different communes and a different year. In their study, most farmers managed to keep the paddy soil wet or with low water level at the time of transplanting in the SRI fields. The difference might have arisen from the difference in the target year between the two studies: 2010 in Ly et al. (2012) and 2014 in this study. As mentioned earlier, the rainfall amount in this district for the wet season started to decline in 2011 and dropped to only 560 mm in 2014, while it was 880 mm in 2010 (Lee and Kobayashi 2017). The large interannual variability in rainfall amount is a common feature in rainfed rice farming (Miyagawa and Kuroda 1988), and it presents a challenge for the interpretation of the results of a single year survey. Evidently, our study has captured the discontinuation of SRI amidst the declining wet season rainfall.
Agronomic determinants of SRI adoption, discontinuation, and rejection Another constraint affecting SRI adoption was the difficulties in transplanting such as the hardships of using the transplanting rope (Fig. 8a). Conventionally, at the study site, rice seedlings are planted at a pace and spacing determined by the individual planters. The planting at regular spacing in SRI is incompatible with the conventional practice, and it is often perceived as being difficult to follow particularly for the planters without SRI experience. It has been reported that, in the initial phase of SRI adoption, farmers depend on family labor, which cannot easily be replaced by hired ones, because of the needs for specific knowledge, training, and experience in Cambodia (Ly et al. 2012) and elsewhere (Noltze et al. 2012). One of the PA farmers at our study site had invented and began to use an apparatus that is self-made from bicycle wheels to make marks on the wet soil surface at a regular spacing very efficiently. The planters can then plant the seedlings at the marks at their own pace without having to wait for the slowest planter, which is the case in transplanting using rope. Despite this seemingly drastic improvement in the labor efficiency in transplanting, no one else has adopted this innovation. It must be noted that the difficulty at transplanting is not limited to the use of transplanting rope for regular spacing. Labor requirement was mentioned by the farmers as a reason against practicing SRI across the adoption categories (Fig. 8b). The PA farmer who invented the apparatus for regular grid planting also mentioned the labor shortage as one of the reasons why he did not practice SRI in some of his fields. A labor peak comes at transplanting, and the farmers in the neighborhood help each other out via labor exchange. The planting method in SRI would increase the labor requirement for the planters who are not familiar with SRI. It must also be noted that SRI dictates the farmer to plant seedlings earlier than the conventional practice and that finding the planting labor at an earlier timing is quite difficult.
13
The labor requirement for transplanting has also been identified as a major contributor to the greater labor demand in SRI by other studies in Cambodia (Ly et al. 2012; Sam and Ouch 2015). A survey on Thai farmers’ perception of SRI practices showed that, of the 66 rice farmers interviewed, nearly half raised the labor intensiveness and the difficulty in transplanting as the constraints to SRI adoption (Doi and Mizoguchi 2013). In Indonesia, secure access to irrigation water and household labor endowment were the two determinants for the continued adoption of SRI (Takahashi 2013). In India also, availability of laborers and irrigation ability was important in determining SRI adoption in addition to the availability of effective information along with other factors such as membership in farmer organizations (Varma 2017). Although previous studies have pointed out that SRI requires more intensive manpower than conventional practices in weed control and water management (Moser and Barrett 2003; Barrett et al. 2004; Ly et al. 2012; Takahashi 2013; Islam et al. 2014; Tumusiime 2017), these were not mentioned by the farmers as reasons for not practicing SRI in this study. Such intensive management may be impossible or even unnecessary in the rainfed environment at the study site.
Economic considerations to SRI adoption, discontinuation, and rejection According to a survey conducted by the SRI-LMB project in 2017, for Cambodian farmers using SRI in 2014–2016 the net return was 225 USD/ha higher compared to those using cultivation conventional practices (SRI-LMB 2017). However, in study area, we found no statistically significant difference in household income between the SRI adoption categories, whereas a stable income from salaried employment improves household liquidity significantly. The increased opportunity of off-farm jobs could reduce the likelihood for a farmer to continue SRI as reported elsewhere (Moser and Barrett 2003; Barrett et al. 2004; Moser and Barrett 2006) and may have contributed to the increasing discontinuation of SRI in recent years (Fig. 2). As more households work in factories, the labor shortage during transplanting would become more severe leading more households to discontinuing or continually rejecting SRI. The increase in discontinuance of SRI (Fig. 2) may therefore have been driven by the increasing opportunity of off-farm jobs in addition to the declining wet season rainfall. The increasing chances of off-farm earnings are also pushing up wages and thereby negatively affecting on-farm livelihood activities. Lately, some farmers conduct transplanting on weekends during the farming season. This causes problems when labor is urgently needed (Ches and Yamaji 2016).
13
Paddy and Water Environment
Conclusions The higher rice yield with lower planting density should be very good reasons for farmers to adopt SRI. They did mention the higher yield and lower seed requirement among the major reasons for the SRI adoption (Fig. 3). However, most of the farmers whom we interviewed continued the rejection of SRI or discontinued SRI, since a majority of them had no water supply in their fields (Fig. 5b). The linkage between supplementary water availability and SRI practices was clearer on the field-to-field basis in Fig 5b of Lee and Kobayashi (2017), where more than 70% of the fields had no availability of supplementary water. Some farmers who had their fields at two or more locations could perform SRI in some of the fields (category PA in Fig. 5), whereas others with fields at only one location and without supplementary water had to discontinue SRI (category DA in Fig. 5). Supplementary water availability in combination with the number of fields thus constitutes one of the major determinants of the farmers decision to adopt or not SRI in their rainfed fields. It follows that a greater and sustainable adoption of SRI at the study site would require increasing the availability of supplementary water from ponds and pumping of river water. The need for small-scale water supply has already been argued for northeast Thailand, where rice production is done mostly in rainfed fields under severe water constraint (Miyagawa and Kuroda 1988; Haefele et al. 2016). Besides the lack of supplementary water, this study has identified difficulties in transplanting according to the SRI principle as another determinant for farmers not to adopt SRI. The regular spacing would become less troublesome as the planters get used to it or adopting an apparatus to facilitate the regular-spaced planting without using a rope. The earlier transplanting that is required is more challenging, however. A labor peak comes at transplanting time, and the recent increase in opportunities to earn income from off-farm jobs has made the labor supply for transplanting difficult, particularly for the earlier planting in SRI compared to the conventional practice. It is noteworthy that the above conclusions have been drawn from two interview surveys on a modest sample size at a specific commune. Accumulation of case studies like this is warranted to draw conclusions of higher generality on the determinants of the farmers’ adoption of SRI across rainfed lowland regions of Cambodia. Acknowledgements The authors gratefully acknowledge the generous supports given by Dr. Makara Ouk, the Director of Cambodian Agricultural Research and Development Institute throughout the study period. This work was supported by JSPS Grant-in-Aid for Scientific Research Grant Number (A) 23255014. The authors’ appreciation is extended to the Korea International Cooperation Agency (KOICA) for facilitating the interview survey by the first author and to the people
Paddy and Water Environment in Popel commune for their cooperation with the interviews. Comments of three anonymous reviewers have contributed to improve the manuscript.
References Adusumilli R, Bhagya Laxmi S (2011) Potential of the system of rice intensification for systemic improvement in rice production and water use: the case of Andhra Pradesh, India. Paddy Water Environ 9:89–97 Balamatti A, Uphoff N (2017) Experience with the system of rice intensification for sustainable rainfed paddy farming systems in India. Agroecol Sustain Food Syst 41:573–587 Barrett CB, Moser CM, McHugh OV, Barison J (2004) Better technology, better plots, or better farmers? Identifying changes in productivity and risk among Malagasy rice farmer. Am J Agric Econ 86:869–888 Ceesay M, Reid WS, Fernandes ECM, Uphoff N (2006) The effects of repeated wetting and drying on lowland rice yield with the System of Rice Intensification. Int J Agric Sustain 4:5–14 Ches S, Yamaji E (2016) Labor requirements of system of rice intensification (SRI) in Cambodia. Paddy Water Environ 14:335–342 Doi R, Mizoguchi M (2013) Feasibility of system of rice intensification practices in natural and socioeconomic contents in Thailand. Int J Sustain Dev World Ecol 20:433–441 Haefele SM, Kato Y, Singh S (2016) Climate ready rice: augmenting drought tolerance with best management practices. Field Crops Res 190:60–69 Islam M, Nath LK, Patel DP, Das A, Munda GC, Samajdar T, Ngachan SV (2014) Productivity and socio-economic impact of system of rice intensification and integrated crop management over conventional methods of rice establishment in eastern Himalayas, India. Paddy Water Environ 12:193–202 Kabir H, Uphoff N (2007) Results of disseminating the system of rice intensification with farmer field school methods in Northern Myanmar. Exp Agric 43:463–476 Lee YH, Kobayashi K (2017) Rice yield increase by the system of rice intensification is dependent on supplementary water availability in rainfed lowland fields of Southern Cambodia. Trop Agric Dev 61:48–55 Ly P, Jensen LS, Bruun TB, Rutz D, de Neergaard A (2012) The system of rice intensification: adapted practices, reported outcomes and their relevance in Cambodia. Agric Syst 113:16–27 MAFF (Ministry of Agriculture, Forestry and Fisheries in Cambodia) (2015) Agricultural Sector Strategic Development Plan 2014–2018. Department of Planning and Statistics Phnom Penh, January 2015 Menete MZL, van Es HM, Brito RML, DeGloria SD, Famba S (2008) Evaluation of system of rice intensification (SRI) component practices and their synergies on salt-affected soils. Field Crops Res 109:34–44 Miyagawa S, Kuroda T (1988) Effects on environmental and technical factors on rice yield in rainfed-fed paddy fields of northeast Thailand. Jpn J Crop Sci 57:773–781 Moser CM, Barrett CB (2003) The disappointing adoption dynamics of a yield-increasing, low external-input technology: the case of SRI in Madagascar. Agric Syst 76:1085–1100
Moser CM, Barrett CB (2006) The complex dynamics of smallholder technology adoption: the case of SRI in Madagascar. Agric Econ 35:373–388 Ndiiri JA, Mati BM, Home PG, Odongo B, Uphoff N (2013) Adoption, constraints and economic returns of paddy rice under the system of rice intensification in Mwea, Kenya. Agric Water Manag 129:44–55 Noltze M, Schwarze S, Qaim M (2012) Understanding the adoption of system technologies in smallholder agriculture: the system of rice intensification (SRI) in Timor Leste. Agric Syst 108:64–73 Sam S, Ouch C (2015) Agricultural technological practices and gaps for climate change adaptation. CDRI Working Paper Series No. 100. Phnom Penh: CDRI Senthilkumar K, Bindraban PS, Thiyagarajan TM, Nde R, Giller KE (2008) Modified rice cultivation in Tamil Nadu, India: yield gains and Farmers’ (lack of) acceptance. Agric Syst 98:82–94 Sinha SK, Talati J (2007) Productivity impacts of the system of rice intensification (SRI): a case study in West Bengal, India. Agric Water Manag 87:55–60 Sothy I (2008) The progress of system of rice intensification in Cambodia-2007. Department of Agronommy and Agricultural Land Improvement (DAALI) in cooperation with Cambodia Center for Study and Development in Agriculture (CEDAC). Phnom PenhMarch 2008 SRI-LMB Project (2013) Background paper Cambodia inception and planning workshop. AIT Pathumthani. http://www.sri-lmb.ait. asia/country/doc/Cambodia_Background_Paper.docx. Accessed 11 May 2018 SRI-LMB Project (2017) Regional review and planning workshop. Hanoi Vietnam AIT Pathumthani, April 23–25. http://www. sri-lmb.ait.asia/countr y/doc/Cambodia%20FPAR%20Prog ress %20Report.pdf. Accessed 11 May 2018 Stoop WA, Adam A, Kassam A (2009) Comparing rice production systems: a challenge for agronomic research and for the dissemination of knowledge-intensive farming practices. Agric Water Manag 96:1491–1501 Styger E, Attaher MA, Guindo H, Ibrahim H, Diaty M, Abba I, Traore M (2011) Application of system of rice intensification practices in the arid environment of the Timbuktu region in Mali. Paddy Water Environ 9:137–144 Takahashi K (2013) The roles of risk and ambiguity in the adoption of the system of rice intensification (SRI): evidence from Indonesia. Food Secur 5:513–524 Tech C (2004) Ecological system of rice intensification (SRI) impact assessment (2001–2003). CEDAC Field Document. http://sri. ciifad.cornell.edu/countr ies/cambodia/camcedacimpact03.pdf. Accessed 11 May 2018 Thakur AK, Kassam A, Stoop WA, Uphoff N (2016) Modifying rice crop management to ease water constrains with increased productivity, environmental benefits, and climate-resilience. Agric Ecosyst Environ 235:101–104 Tumusiime E (2017) Suitable for whom? The case of system of rice intensification in Tanzania. J Agric Educ Ext 23:335–350 Varma P (2017) Adoption of system of rice intensification under information constraints: an analysis for India. J Dev Stud. https://doi. org/10.1080/00220388.2017.1336541 Yang SK (2002) Ecological system of rice intensification (SRI) in Cambodia results and field experience in 2000–2001. CEDAC field document
13