Narrowing the rice yield gap in East and Southern Africa: Using and adapting existing technologies

Agricultural Systems 131 (2014) 45–55

Contents lists available at ScienceDirect

Agricultural Systems journal homepage: www.elsevier.com/locate/agsy

Review

Narrowing the rice yield gap in East and Southern Africa: Using and adapting existing technologies Nhamo Nhamo ⇑, Jonne Rodenburg, Negussie Zenna, Godswill Makombe 1, Ashura Luzi-Kihupi 2 Africa Rice Center, East and Southern Africa, P O Box 33581, Dar es Salaam, Tanzania

a r t i c l e

i n f o

Article history: Received 8 March 2012 Received in revised form 4 August 2014 Accepted 5 August 2014

Keywords: Good Agricultural Practice (GAP) Soil degradation Cultivars Systems Innovations Africa

a b s t r a c t The importance of rice production in sub-Saharan Africa (SSA) has significantly increased over the past decades. Currently, rice plays a pivotal role in improving household food security and national economies in SSA. However, current rice productivity of smallholder farms is low due to a myriad of production constraints and suboptimal production methods, while future productivity is threatened by climate change, water shortage and soil degradation. Improved rice cultivars and agronomic management techniques, to enhance nutrient and water availability and use efficiencies and to control weeds, have the potential to increase yields. The aim of this study was to assess the relative contribution of such technologies to enhanced rice productivity. Relative yield gains emanating from nutrient, water and weed management were surveyed and calculated from literature. Partial budgeting was used to evaluate viability of fertilizer technology under GAP. Substantial yield gains ranging from 0.5 t ha1 to 2.9 t ha1 are projected following the use of improved technologies. Relative yield gains decreased in the following order: weed management (91.6%) > organic fertilizer application (90.4%) > bunding (86.7%) > mineral fertilizer application (51.9%) > tied ridges (42.6%). Combining fertilizer with unimproved rice cultivars led to negative returns. The lack of integration of improved technologies, to increase synergies and alleviate socio-economic constraints, largely explained the existing yield gaps. The gains obtained through improved rice cultivars can be further enhanced through application of Good Agricultural Practices (GAP), improving nutrient, water and weed management technologies, based on the local resource availabilities of small farms. We therefore propose adapting technologies to local conditions and developing and using rice production decision tools based on GAP to enable rice farmers in SSA to improve resource-use efficiencies and crop productivity at the farm level. Ó 2014 Elsevier Ltd. All rights reserved.

Contents 1. 2. 3.

4.

5.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yield gain and economic analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rice yield responses to improved technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Soil fertility management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Water management. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Weed management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Improved varieties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Narrowing yield gaps through integrated management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Crop management systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Local adaptations and holistic approaches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Management tools for improved resource use in rice systems in ESA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Implications for research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

46 47 47 47 48 48 49 50 51 51 51 53

⇑ Corresponding author. Present address: International Institute of Tropical Agriculture (IITA), P O Box 310142, Chelston Lusaka, Zambia. Tel.: +260 211 840365. E-mail addresses: [email protected], [email protected] (N. Nhamo), [email protected] (J. Rodenburg), [email protected] (N. Zenna), [email protected], [email protected] (G. Makombe), [email protected] (A. Luzi-Kihupi). 1 Present address: University of Limpopo, Turfloop Graduate School of Leadership, Box 756, Fauna Park 0787, South Africa. 2 Present address: Sokoine University of Agriculture (SUA), P O Box 77, Morogoro, Tanzania. http://dx.doi.org/10.1016/j.agsy.2014.08.003 0308-521X/Ó 2014 Elsevier Ltd. All rights reserved.

46

N. Nhamo et al. / Agricultural Systems 131 (2014) 45–55

6.

Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

1. Introduction The importance of rice as a commodity has significantly increased over the past decades in Sub-Saharan Africa (SSA) (Seck et al., 2010). Rice plays a pivotal role in African rural household food security and national economies. Since the 1990s urbanization and increased income associated with rural–urban migration (Kennedy and Reardon, 1994) has led to an increase in per capita rice consumption. In SSA rice grain yield per unit area and the total area under production have stagnated (Otsuka and Kalirajan, 2005). There are however still possibilities to expand area under rice and improve productivity given the positive land balance (FAO, 2010) and the relatively low level of adoption of modern technologies (Balasubramanian et al., 2007). Clearly, there is a need to study important yield reducing factors closely in order to determine strategies to help increase and maintain rice productivity on farmers’ fields and, through that, overall regional rice production. In East and Southern Africa (ESA), Madagascar and Tanzania are the major rice producing countries (Table 1), while Rwanda is the smallest producer (FAO, 2010; Kanyeka et al., 1996; Rodenburg and Demont, 2009). In terms of area under rice, the rain-fed lowlands are the dominant ecosystems in ESA, comprising 55% of the total area. Irrigated rice ecology (both highland and lowland) comprises 27% while rain-fed uplands comprise 18% of the area under rice. Though there are disparities across countries in the region, the biophysical conditions in ESA (topography, water reservoirs, rainfall distributions and soils), suggests that there is untapped potential for improving rice production. Ferralsols, Acrisols, Arenosols, Nitosols and Lixisols are the dominant soil types found in the ESA region (Bationo et al., 2006; Bekunda et al., 2002; Hartemink, 1997; Nandwa and Bekunda, 1998). Due to erosion and degradation, soils on uplands are relatively less fertile and more acidic than those on lower positions on the catena, with the latter being accumulation zones for soil mineral sediments, nutrients, organic matter and (rain or run-off) water (e.g. Andriesse et al., 1994; van der Heyden and New, 2003). It is for this reason that there exist a relatively large agricultural potential in rain-fed lowland systems (inland valleys, also known as mbuga in East Africa and vleis, dambos, mapani or matoro,

and inuta or amaxhaphozi in Southern Africa according to Acres et al. (1985)) in particular for rice production (e.g., Rodenburg et al., in preparation). However, the lowland ecosystems should not be developed indiscriminately for the sole purpose of agricultural production, as they are often fragile or harbour a range of natural resources (e.g. biodiversity) linked to important ecosystems functions worthy of conservation (e.g., McCartney and HoughtonCarr, 2009; Sakane et al., 2011; Verhoeven and Setter, 2010). There is a growing realization that rice production is important for advancing the agricultural contribution to the national GDP (e.g., Seck et al., 2012). In Madagascar for instance, rice is the main staple food crop and an important export commodity (Garenne, 2002), while Kenya, Mozambique and Uganda are net importers of rice (NPA, 2007). Improving domestic production can reduce imports. If production increases alongside quality, for instance through investments in post-harvest grain-quality infrastructure, it will augment the market share of locally produced rice (e.g., Demont and Rizzotto, 2012). However, major constraints to rice production are of biophysical (i.e. soil nutrient depletion, weed infestation, variable rainfall patterns, low and under-developed irrigation infrastructure), socio-economic, institutional and political (i.e. lack of financial resources, labour shortages, low levels of education, weak infrastructure, lack of conducive policies)nature. Solving these constraints could bridge the existing large gap between current farm level production and the potential production. The term ‘yield gap’ is used to indicate the difference between the biological and climatic potential yield and the average actual crop yield produced by farmers (Lobell et al., 2009). Factors affecting crop growth and development are radiation and temperature (yield determining), water and nutrition (yield limiting); the attainable yield is the potential yield limited by these two factors in a given environment (Rabbinge, 1993). An additional factor affecting crop growth is pest and diseases (yield reducing). In addition, productivity is also determined by factors such as cultivar choice and crop management. The interaction between the above factors determines the actual yield level at a particular location. In irrigated areas productivity is primarily determined by radiation and temperature whereas in rain-fed areas, precipitation and soil moisture storage capacity are important factors (De Wit, 1992).

Table 1 Harvested area under cultivation (ha) and mean yields (kg ha1) for rice under rainfed upland (RU), rainfed lowland (RL) and irrigated ecosystems from 8 countries in East and Southern Africa. Country

RU

RL

IR

Total Area (ha)  1000

Rice yield (kg ha1)

Burundi Kenya Madagascar Malawi Mozambique Rwanda Tanzania Uganda

4 0 29 0 39 0 23 45

74 0 18 72 59 92 73 53

21 100 52 28 2 8 4 2

21 19 1300 53 204 10 665 119

3310 3570 2770 1740 960 4400 1860 1360

Regional share (%) Total area under rice Average yield (kg ha1) Standard deviation of the mean Difference in yields (kg ha1)

17.5

55.3

27.2

– 2391

–

FAO (2010).

2496 1204 3440

N. Nhamo et al. / Agricultural Systems 131 (2014) 45–55

The main objectives of this paper are to determine rice yield gains resulting from use of improved technologies and to analyse how technical interventions should be adapted to narrow the gap between potential and actual rice yields. We hypothesize that agronomic practices and socio-economic conditions are highly interdependent and should therefore be handled in a holistic manner in order to attain increased yields, and that attention should be given to synergies resulting from applying technologies that have the potential of addressing two or more yield reducing or limiting factors. 2. Yield gain and economic analysis Published data from over 105 relevant journal articles and FAOSTAT were analysed for yield differences. To determine the effects of improved cultivars and management practices on yield, yield gains were calculated using Eq. (1). 1

YG ðkg ha Þ ¼ Y T  Y 0

ð1Þ

where YG is yield gain; YT is the yield following a treatment effect (e.g. fertilizer application, weed management) and Y0 is yield without any treatment (from negative control plots). In order to compare the effects of technologies on yields across different sites we calculated a relative yield gain (RYG) from each experiment expressed as a percentage using Eq. (2).

RYG ð%Þ ¼ 100  YG=Y 0

ð2Þ

Partial budgeting was used to evaluate the economic viability of component technologies. 3. Rice yield responses to improved technologies 3.1. Soil fertility management Nutrient depletion through crop and residue removal and soil degradation (erosion)is a major threat to rice production in ESA (Hartemink, 1997; Hartemink et al., 2005). Calculations showed that as high as 70% N, 80% P and 63% K can be lost by soil erosion especially on fields with more than 5 % slope (Hartemink et al., 2005). Rice production without soil fertility amendments, a common practice on smallholder farms in ESA, is consummate to high rates of soil nutrient depletion (Mghase et al., 2010). Low application rates of poor quality soil fertility inputs cannot halt nutrient mining leading to reduced returns from land, water, labour (e.g., for weed management) or any other necessary resources and inputs. While mineral and organic fertilizers are important nutrients sources, little is known about the effectiveness of combinations and formulations

47

used by smallholder famers on rice yields. Blanket application of variable combinations of N, P and K are widely acknowledged in ESA (Nandwa and Bekunda, 1998) but smallholder rice farmers hardly use these recommendations. Due to financial constraints, farmers often apply reduced amounts of fertilizer (Mwaseba et al., 2006). Inappropriate application of fertilizers can become an inefficient and unprofitable practice (Poulton et al., 2006). There is a need for locally adaptable, and balanced recommendations suitable for a range of crop environments. This would require a thorough understanding of the specific system (Bekunda et al., 2002; Meertens et al., 1999). In addition, farmers need access to credits and markets in order to be able to follow-up on recommendations. Based on our analyses of on-farm studies carried out by Kaihura et al. (1999), Kajiru et al. (1998), Kayeke et al. (2007) and Meertens et al. (2003), a relative yield gain of 52%, equivalent to 948 kg ha1 of rice grain can be obtained from the use of mineral fertilizer (Table 2). A closer analysis showed that the initial linear yield gains resulting from N application start to gradually decrease above 60 kg N ha1 (Kaihura et al., 1999; Kajiru et al., 1998; Kayeke et al., 2007; Meertens et al., 2003). This result suggests increased losses occur under higher fertilizer application rates probably because of poor nutrient recovery by the crop and traditional fertilizer management practices. Improved fertilizer management practices have been used successfully to increase fertilizer use efficiency however, arriving at the right quantities and timing of application in relation to the crop phenology remains a challenge to most subsistence farmers in ESA (Esilaba et al., 2005). In ESA average fertilizer application rates on cereals in general range from 5 to 20 kg N ha1 (Nandwa and Bekunda, 1998; Nhamo et al., 2002; Okalebo et al., 2006) and similar rates are commonly observed in rice in particular (Kajisa and Payongayong, 2011; Nakano and Kajisa, 2012). Subsidies, improved agro-dealer networks and re-packaging of fertilizers into smaller portions are measures used to improve fertilizer use by resource poor farmers (Poulton et al., 2006). The success of such interventions has however been low. Meertens and Roling (2000) reported low adoption of fertilizer by rice farmers due to suboptimal availability of urea at the local markets and a lack of incentives due to high fertilizer prices. Farmers are least likely to embrace fertilizer use if they lack information on yield benefits, ease of implementation, profitability and compatibility to their farming systems (Meertens and Roling, 2000). Contrary to locally adapted advices, general fertilizer recommendations fail to capture the within-plot and within-farm variability which are often larger than between-farm differences (Mafongoya et al., 2006). Piha (1993)suggested that within-season management of fertility input can raise the economics of fertilizer use and reduce the risk of its application in areas prone to mid-season droughts. By applying P, K and S fertilizer as basal

Table 2 Yield gains (kg) characteristics from experimental data (YT–Y0) from published journal article on nutrient, water, weed management trials on rice cultivars calculated as the difference between each treatment mean and the control. Technology

Mean Yield gain (kg)

Maximum gain (kg)

Source(s)

Nutrient management Mineral Organic

948 (51.9) 1424 (90.4)

2931 (329) 2830 (196)

Kaihura et al. (1999), Kajiru et al. (1998), Kayeke et al. (2007), Meertens et al. (2003) Kaihura et al. (1999), Kajiru et al. (1998), Kayeke et al. (2007), Kijima et al. (2006), Makoye and Winge (1996), Meertens et al. (2003), Menete et al. (2008), Otsyina et al. (1995)

Water management Bunds Tied ridges

500 (86.7) 1162 (42.6)

700 (180) 2931 (149)

Raes et al. (2007) Kaihura et al. (1999)

1241 (91.6)

2600 (182)

Kayeke et al. (2007), Kijima et al. (2006)

Weed management Cultural and mechanical methods

Relative yield gain (%) in parenthesis.

48

N. Nhamo et al. / Agricultural Systems 131 (2014) 45–55

applications while varying N applications depending on rainfall patterns yields increased by 25–42% and profits by 20–40%. Combining such an approach together with simple but effective field assessment techniques for farmers, such as the leaf colour charts, can substantially improve the N fertilizer efficiency on rice. Organic fertilizers address the biological, chemical and physical fertility of soil leading to long-term increased availability of nutrient and water to the crop (Okalebo et al., 2006). The successful use of compost in the System of Rice Intensification (SRI) leading to yields of around 9 t ha1 shows the potential of organic fertilizers in lowland rice (Moser and Barrett, 2003;Stoop et al., 2009). However, to obtain such high yields with this system, compost or manure application of 13 (Styger et al., 2011), 20 (Abo et al., 1998; Kaihura et al., 1999) and even 35 t ha1 (Tsujimoto et al., 2009) were reportedly used or recommended. Such application rates are often far more than what an average livestock farmer produces annually. The challenge of using organic amendments on small farms is often the supply of such large required to supply balanced nutrition to crops (Mafongoya et al., 2006). Our analyses showed that organic fertilizer application result in higher rice yield gains than the application of mineral fertilizer (Table 2), and this practice also has a residual effect. The organic inputs used in the analysed studies include animal manure, banana residues, spear grass, elephant grass, kitchen waste, maize stover, potato peels, cassava peels, coffee husks, household refuse after composting, bean trash, and biomass of Tithonia diversifolia, Calliandra calothyrsus and Leucaena leucocephala (e.g., Bekunda and Woomer, 1996; Meertens et al., 1999; Nhamo et al., 2002; Okalebo et al., 2006; Sseguya et al., 1999). Besides sole application of organic materials the combined application of organic and mineral fertilizers have resulted in higher yields (Alemu and Bayu, 2005). Sistani et al. (1998) demonstrated an increase in rice yield when both rice straw and rice hull ash were combined with mineral N and P fertilizers. The potential of building and maintaining soil nutrient stocks using organic fertilizers (e.g., residues, compost and manure) remains however elusive as often farmers prefer to apply them to higher value crops (Krupnik et al., 2012a–c) or the material is needed to meet other needs (e.g., building materials and fodder). Further, the decomposition rates associated with tropical soils are generally high (Mafongoya et al., 2006; Nkonya et al., 2005). Fallowing used to be a predominant practice of replenishing soil fertility on cultivated soils under traditional rain-fed farming systems (Bationo et al., 2006; Okalebo et al., 2006). Given limited access to land due to population growth, fallowing to improve soil nutrient stocks is currently less frequently used (e.g., Demont et al., 2007). However, there is potential for improved fallows where multipurpose legumes are used to enhance soil fertility, in particular through biological nitrogen fixation (Giller, 2001). In upland rice production systems the use of improved fallows and relay intercropping with legumes has been investigated in West Africa (e.g., Becker and Johnson, 1999). The use of Cajanus cajan as relay intercrop and improved fallow species showed to have a positive effect on yields of the following rice crop (Akanvou et al., 2002). Legume fallows have the potential to play an important role in intensified upland rice systems, but effective implementation of this technology requires more research on the most effective short-duration multipurpose legume species and the most appropriate management (Becker and Johnson, 1999). Loevinsohn et al. (1994) reported variable success when farmer groups, in cooperatives and associations experimented on the use of Sesbania spp. to improve the fertility status of the soils in inland valley rice systems in Rwanda. 3.2. Water management Demand for water is driven by competing uses including domestic consumption, crop, livestock and industrial production,

as well as fisheries and maintenance of wetland ecosystems. Climate projections suggest that rainfall variability will increase, higher rainfall amounts are expected in East Africa whereas drought incidences will be more frequent in the Southern and Western parts of Africa (Giannin et al., 2008). In ESA irrigation systems are often underdeveloped or sub-optimally used (Mulwafu and Nkhoma, 2002) while rain-fed rice production is more vulnerable to rainfall irregularities and uncertainties. To reduce water stress in crops, rainwater harvesting techniques need to be promoted. Using data from our synthesis, the relative yield gain (RYG) emanating from the use of bunding (86.7%) was double that from tied ridges (42.6%) (Table 2), though higher maximum gains were obtained in the latter. Bunds and tied ridges can alleviate moisture stress during drought in rain-fed rice systems (Raes et al., 2007). However, lack of awareness and the labour involved in bund construction could explain the low adoption of such practices as observed in Tanzania (Raes et al., 2007). Obviously, the positive effects of temporary and permanent bunds, ridges and furrows on yields are more pronounced during years with moderate droughts (Singh, 2006). Luzi-Kihupi et al. (2004) highlighted the importance of bunding, on-farm reservoirs, irrigation and drainage systems in controlling water. Where water management structures are poor the risk of yield loss due to drought, weeds and nutrient losses increases. The risk of soil erosion followed by soil degradation and reductions in water holding capacities, as observed by Kaihura et al. (1996), highlights the importance of controlling surface water movement on crop fields. Socio-economic and agronomic limitations on many smallholder farms are often reflected in poor water management (Dobermann, 2004). Combining water management and nutrient management in rain-fed lowlands, through bunding, proved to increase yields (e.g., Becker and Johnson, 2001; Toure et al., 2009). While manual bund construction allows for timely planting, it requires a lot of manual labour, a scarce resource in subsistence farms in Africa, in particular during the planting period (Meertens and Roling, 2000). 3.3. Weed management Rodenburg and Johnson (2009) conservatively estimated the cost of weed-inflicted yield losses in rice, despite control and exclusive of the costs for weeding operations, to be $ 1.5 billion for sub-Saharan Africa. The high labour demand for weed control has been generally acknowledged as a major problem in rice production in SSA (Balasubramanian et al., 2007; Lawrence et al., 1997). However, according to our analyses labour inputs focussing on weed management resulted in high yield gains in East Africa (Table 2). Although based on few studies, our analysis shows that weed management technologies account for the highest yield gains in rice production in ESA (Table 2). Improved weed control practices in rice resulted in a RYG of 91.6% (equivalent to 1.241 kg ha1 of paddy) compared to farmers’ practices (Table 2). However, where weed management is not adequate, yield losses as high as 250 kg ha1 of paddy occur and in extreme cases a total crop failure. This result indirectly corroborates reports that weeds together with birds are major problems faced by rice farmers (Seck et al., 2012). Common weeds in rice in ESA are: Echinochloa colona, Sphenoclea zeylanica, Monochoria vaginalis, Hygrophila spinosa, Commelina benghalensis, Corchorus olitorius, Ludwigia hyssopifolia, wild rice (Oryza longistaminata and Oryza punctata) and different species of the Cyperaceae family (Meertens et al., 1999). Rodenburg et al. (2010) has highlighted the increasing importance of parasitic weeds (e.g. Striga spp. and Rhamphicarpa fistulosa) in rice in subSaharan Africa, reporting Striga asiatica problems in Tanzania and Madagascar and Striga hermonthica in Kenya and Uganda.

49

N. Nhamo et al. / Agricultural Systems 131 (2014) 45–55

Labour intensive and time consuming manual methods of controlling weeds (i.e. hand pulling or the use of the short-handled hand hoe) are often used to clean rice fields of weeds (Rodenburg and Johnson, 2009). Furthermore, cultural measures that are frequently observed in subsistence rice production systems in SSA and that contribute to weed control include land preparation (e.g. dry tillage, deep tillage, levelling), transplanting, flooding and crop rotations with non-cereal crops (AmpongNyarko and De Datta, 1991; Zimdhal, 2007). On smallholder farms serious labour shortages occur making manual weeding highly inappropriate for improved rice production. The labour situation has been exacerbated by rural–urban or rural–rural migration for non-agricultural employment and the effects of the HIV–AIDS pandemic on rural communities. Data from Southern Africa showing low adoption of labour intensive agricultural technologies corroborate this position (Perret and Stevens, 2006). In order to reduce labour demands, interventions such as (small scale) mechanization and use of chemical weed control methods, where possible, are needed. No single weed control methods provide an effective solution to all weed species at all times, suggesting that it is important to develop an integrated approach to weed control (Rodenburg and Johnson, 2009). Further, cultural weed control methods promoted in such an integrated approach (e.g. levelling, flooding, transplanting, use of clean rice seed, high and uniform plant densities) also benefit the crop productivity in general and therefore support a sustainable rice production system. Improving soil fertility technologies can lead to Striga control. Examples from Tanzania, for instance, show that combinations of green manures and mineral fertilizers can reduce S. asiatica infestation while at the same time increasing N supply to rice (Kayeke et al., 2007; Riches et al., 2005). Crop rotations, green manures or improved fallows and intercropping are widely known for controlling weeds in general and parasitic weeds in particular through (light) competition or suicidal germination (Rodenburg et al., 2010). Such crop combinations also increase resource use efficiency, break disease cycles thereby reducing risk of crop failure. Reasons for low adoption of this type of cropping systems technologies are high labour requirements for clearing and incorporating into the soil and the general lack of legumes that fit in existing farming systems (Becker et al., 1995). Legumes that combine an economic or food value with weed control and soil improvement capabilities stand a better chance of adoption (Becker and Johnson, 1999). The control of weeds through the use of high yielding and weed-competitive cultivars is another potential technology. A range of newly developed lowland NERICA cultivars were identified to be weed competitive and high yielding in studies done in West Africa (Rodenburg et al., 2009), while among the NERICA cultivars adapted to uplands, effective mechanisms of resistance against S. hermonthica (Cissoko et al., 2011; Jamil et al., 2011) and S. asiatica (Cissoko et al., 2011) have been identified. Such improved cultivars could also play a key role in integrated weed management in rice in ESA. The challenge in using this approach is how to deal with wild rice species such as O. longistaminata and O. punctata. These widely distributed wild rice species are highly competitive and cannot be easily controlled through herbicides (Munene et al., 2008) or using mechanical weed control, and such species are also predicted to become a more persistent problem given projected increases in atmospheric CO2 concentrations (Rodenburg et al., 2011).

ern varieties (IMV) generally have higher yield potential than traditional varieties and a higher yield response to fertilizer application. The yield gains reported in Table 2 show the effect of plant nutrient use, water and weed management practices on modern improved cultivars in ESA. However, in Mozambique, Tanzania and Uganda there has been slow progress in the adoption of IMVs as farmers still grow traditional landraces with low responsiveness to inputs (Kijima et al., 2011; Meertens et al., 1999; Menete et al., 2008). Indeed, there are major constraints in the dissemination and adoption of IMVs as they do often not possess some farmers’ and consumers’ preferred qualities that traditional cultivars have. Farmers in ESA, often growing for their own consumption, value cooking quality and taste, adaptability to the local environment and the ability to yield under minimum management. For cultivars combining such highly valued traits, farmers could accept some of the negative associated characteristics such as lower yield potential, later maturity, lodging and poor milling quality (Meertens et al., 1999). Aroma and softness after cooking were found to be important parameters for farmer adoption (Luzi-Kihupi et al., 2007). These two characteristics are found in Supa India (locally known as ‘Kilombero’) in Tanzania, Faya from Malawi, Sindano from Kenya and Sokotera from Zanzibar (Meertens et al., 1999). Indeed, Mwaseba et al. (2006) noted the importance of including cooking quality preferences of farmers in breeding IMVs and its potential positive implications on adoption. On the other hand, traditional varieties often have long crop cycles, increasing the risk of crop failure in the event of a shortened season. Work of Mwaseba et al. (2006) described the economic advantage of IMVs especially when they are harvested for sale during the period when the long duration traditional cultivars are not yet mature. An effort to improve the popular traditional variety Supa India in Tanzania, resulted in varieties that indeed combine the desired grain qualities with early maturing, high(er) disease resistance and high(er) responsiveness to water and fertilizer inputs (Kanyeka et al., 1996; Luzi-Kihupi et al., 2009). The two popular Supa India derivatives are SARO 5 (TXD 306) and Mwangaza. SARO 5 has high potential under irrigated conditions whereas the shortduration Mwangaza is suitable for (rain-fed) lowland as well as upland conditions. The major bottle neck that may hamper the wide adoption of SARO5 is its lack of resistance and tolerance to rice yellow mottle virus (RYMV) according to (Msomba et al., 2004). RYMV has been found to be highly infectious and destructive and threatens rice production throughout Africa (Abo et al., 1998). The application of cultivar characteristic (of landraces) to solve local challenges of water scarcity and bird attack has been reported by Meertens et al. (1999). They observed that Supa India

3.4. Improved varieties

Source of paddy yield data: Meertens et al. (2003). Post harvest price of paddy based on key informants from Mvomero district, Tanzania. c Price of 50 kg bag of Urea (46% N) was 45,000 Tanzanian Shilling (around 25–30 USD). d Exchange rate: 1 USD = Tsh 1560.

There is no doubt that the use of improved rice varieties account for a large proportion of the yield gains obtained in the past decade in Africa (Renkow and Byerlee, 2010). Improved mod-

Table 3 The economic benefit of fertilizer use on rice: an example of yield gains (the difference between fertilizer treatment and non-fertilized control), cost of fertilizer application and the net benefits of this technology from studies carried out in Tanzania.

a

b

Item

Amount

Yield gain following fertilizer application (kg ha1) Price of paddy (TS kg1) Gross benefit (TS) Fertilizer price (TS kg1) Total cost of fertilizer (TS ha1) Net benefit (TS ha1) M.R.R (%)

730a 600b 438,000 1957c 58,696 379,304d 646

50

N. Nhamo et al. / Agricultural Systems 131 (2014) 45–55

(Kilombero) was grown in water limited conditions, while high water demanding Tondogoso and Lugata were popular in the valley bottoms and Kahogo and Sindano with awns on the spikelets (considered less susceptible to bird attack) were used to mitigate losses from bird attack. While breeding for higher yields has been a noble and worthwhile objective for many years, current challenges faced by farmers require breeding programs to include many other attributes such as weed competitiveness, disease and parasitic weed resistance and grain quality including aroma, milling and cooking quality. Resource use efficiency of both rain-fed and irrigated rice cultivars is becoming increasingly important given the level of degradation on productive soils (Bationo et al., 2006; Nandwa and Bekunda, 1998), and the cost of inputs required to produce rice. NERICA cultivars that combine good adaptation with short cycles, weed competitiveness and resistance to important diseases (except RYMV) are examples of cultivars that are capable of producing under a range of yield reducing factors (e.g., Balasubramanian et al., 2007; Seck et al., 2012; Wopereis et al., 2008). One additional next challenge is for IMV to be able to cope with climate irregularities, following climate changes (e.g., Jagadish et al., 2012). To combat stresses attributed to climate change cultivars are being developed that maintain productive capacities under (temporary) temperature or water stress (Brar and Virk, 2010). This development is worth consideration as an investment that can benefit ESA. A good response to external inputs and management is one of the valuable characteristic of IMVs. Table 3 shows high marginal rates of return following the application of 30 kg ha1 on responsive rice cultivars. In this example, even after considering the downside risk (e.g. one standard deviation) the practice is still profitable. However, a more complete fertilizer response function would be required to estimate the economically optimum fertilizer application rate. Furthermore this example (Table 3) also highlights the importance of combining soil fertility technologies with improved rice cultivars as illustrated in Fig. 1. The local demand for rice has been on an increase due to improved incomes, organized markets (local level) and urbanization. Work of Luzi-Kihupi et al., 2007; Meertens et al., 1999; have highlighted the importance of the preference to local rice which

Mineral + organic fertilizer, soil reserves NUE/AUE – timing , recovery Indigenous fertility status Topography/Catenal position

Site-specific nutrient management

• • • •

Water-saving systems Improved WUE Moisture conservation techniques Timing of flooding

has attributes such as aroma, grain shape, size and structure. Such preference has created a nitch in the local market and hence protects local producers from competing directly with external rice sources where subsidies could be available for farmers. Work of Msomba et al., 2004; Mwaseba et al., 2006; Kanyeka et al., 1996; Luzi-Kihupi et al., 2009; Shayo et al., 2006; Tsujimoto et al., 2009) in Madagascar, Tanzania, Uganda and Zanzibar highlighted the consumer preference to locally produced rice in ESA. However, the development of these market and the response to global dynamics of trade in crop commodities e.g., rice needs to be studied in order to identify changes in trends over time. As noted by Rugumamu (2014), future exploitation of both domestic and international markets may result from increased support on rice production in the smallholder farming sector. 4. Narrowing yield gaps through integrated management Tran (2004) and Lobell et al. (2009) clearly differentiate between yield gap components. Large yield gaps exist in irrigated rice systems; in Madagascar for instance, a gap of 2.1 t ha1 (a difference between 6.1 and 4.0 t ha1) was determined in irrigated rice (Tran, 2004). The real challenge is to narrow the gap between farm level yield and the potential yield and to identify the main drivers of yield gains. We propose a focus on nutrient, water and weed management as the main biophysical factors and labour shortages, lack of knowledge and weak financial position of the farmers as the socio-economic factors, and analyse this in the context of rice farming in ESA. The figures on production potential using components of Good Agricultural Practice (GAP) from published experimental data are known (this publication) however the farm level yield from improved varieties are not readily available limiting our discussion of the yield gap. Previous sections have focused on identifying yield gains that can be obtained by addressing single constraints, while many constraints co-exist (e.g. poor soil fertility and parasitic weeds, poor water control and weed infestation) and many of the technological solutions target multiple constraints (e.g. bunding improved water, weed and nutrient management). The next step is to examine integrated and locally-adapted approaches as our hypothesis is that using such an approach will result in yield benefits beyond those that can be expected from single-target approaches. Here we will

Improved adapted varieties

GAP Optimized water management

HYV Drought tolerant Disease, pests resistant/tolerant Weed competitive Nutrient efficient Aromatic

Locally adapted integrated weed management

Crop sequences Rotating control practices WC/WS cultivars Timing of interventions Improved crop establishment

Fig. 1. A schematic presentation illustrating linkages and attributes of four component technologies (nutrient, water, weed and varietal management), which can be used as a basis for formulating Good Agricultural Practices (GAP) for rice systems in East and Southern Africa. HYV – high yielding varieties; NUE – nutrient use efficiency; AUE – agronomic nutrient use efficiency; WUE – water use efficiency; WC – weed competitive varieties; WS – weed suppressing varieties.

N. Nhamo et al. / Agricultural Systems 131 (2014) 45–55

give a short overview of integrated management practices used previously in rice in ESA and guidelines on how to build on these previous experiences, coming to local adaptations of these practices. 4.1. Crop management systems Different crop management systems are developed for rice. Here we will discuss three prominent ones, as much as possible in the context of rice-based systems in East and Southern Africa, to illustrate the diversity in objectives and approaches and to show how such systems put the aforementioned yield gain hypotheses into practice. For rain-fed and irrigated lowlands, Integrated Crop Management (ICM), for rice systems also referred to as Integrated Rice Management (IRM), is developed (Donovan et al., 1999; Haefele et al., 2000, 2001, 2003; Kebbeh and Miezan, 2003; Wopereis et al., 1999). IRM comprises recommendations ranging from land preparation to post-harvest management and uses a cropping calendar with 5 main crop stages (Table 4). Transplanting of seedlings in the 3–4 leaf stage and at optimal spacing (20  20 cm) is applied together with improved water, nutrient and (timing of) weed management. In irrigated lowlands, with full control over water, the System of Rice Intensification (SRI) can be applied (Stoop et al., 2002). It is composed of a set of management components; transplanting of single and young (14-day old) seedlings at a wide plant spacing (at least 25  25 cm), alternate wetting and drying (AWD) water management, mechanical weeding (using a rotary weeder) and the use of large amounts of (preferably organic)soil fertility amendments. Outside Madagascar, SRI has been tested in some part of ESA (e.g., Menete et al., 2008) and in West Africa (e.g., Krupnik et al., 2012a–c; Styger et al., 2011) with mixed results. The emphasis on seedling age and its relationship with overall yield has been a point of debate concerning the effectiveness and applicability across cultivars grown under different conditions. Other debatable components are the use of large amounts of organic soil amendments as outlined above, and the feasibility of alternate wetting and drying for smallholder farmers. Recent work by Krupnik et al. (2012a) has shown that adaptations of SRI, for instance replacing unavailable compost by available rice straw, and the involvement of farmers in developing such adaptations, hold a promise to improving such production practices. In rain-fed upland systems, Conservation Agriculture (CA) is a crop management system applied on vulnerable soils (Scopel et al., 2012). Here tillage is minimized, and the soil is protected by mulch derived from rotation or intercrops (crop residues). Some successes of CA in rice in Africa are reported from work in the vulnerable highlands of Madagascar (e.g., Scopel et al., 2012) but the applicability as well as the adoption potential are rather localised. The knowledge and labour requirements of all these systems can be a serious hindrance to adoption by farmers (Kebbeh and Miezan, 2003; Krupnik et al., 2012a; Moser and Barrett, 2003; Scopel et al., 2012). However, despite the differences and limitations of each of the three cropping systems (Table 4) they are all in principle built on Good Agricultural Practices (GAP). Local adaptations of such systems or the adoption of one or a few of their components at a specific location might provide farmers with some advantages in terms of yield or resource-use efficiency. 4.2. Local adaptations and holistic approaches Narrowing yield gaps suggests refinement of available technologies, making them more efficient and suitable under the prevailing local socio-economic and biophysical circumstances thus reducing environmental damage and negative returns to invest-

51

ments on inputs. The development and use of site-specific recommendations is imminent in order for smallholder farmers to obtain yield levels that optimize labour, land and applied inputs. Developing Good Agricultural Practices (GAP) requires knowledge of the available technologies, farmers and farm characteristics, and access to markets. Fig. 1 illustrates the linkages between technologies and how they can contribute to GAP. The GAP approach recognizes the importance of addressing research questions within each component technology and emphasizes the resultant synergies between component technologies within GAP. Local adaptation of technologies is of key importance in GAP. Following this approach, technology packages, such as the System of Rice Intensification (SRI) or Conservation Agriculture (CA) should only be introduced where they are likely to be value-adding, and they should be adapted to the local conditions, rather than treated as fixed packages (e.g., Krupnik et al., 2012a). Conservation agriculture, has gained popularity in ESA on many cereals because it addresses firstly the soil degradation (long-term) and water losses (short-term) that occur on vulnerable rain-fed uplands. Secondly, conservation agriculture has the potential to improve soil quality (soil health). However, the application of conservation agriculture requires careful consideration of the farm and farmer circumstances to ensure its relevance and feasibility (Giller et al., 2009) and the suitability of conservation agriculture for (upland) rice has not been thoroughly examined. Most agricultural technologies are only sporadically adopted by farmers (e.g., Balasubramanian et al., 2007). Therefore we concur with Ducrot and Capillon (2004) who propose, in order to increase adoption rates, that introductions of new, or adapted, technologies need to be combined with promotion or training of the associated organizational and management skills. 4.3. Management tools for improved resource use in rice systems in ESA We observe roughly two groups of farmers in ESA: (1) the traditional rice growers that often grown rice for their own consumption (subsistence farming) and (2) the new farmers who grow rice as a cash crop (i.e., for profit). Traditional rice growers face challenges in changing their production practices to those required for improved cultivars while the new rice farmers have little to no experience with rice production, e.g. the case of farmers in Uganda (Meertens et al., 1999). Both groups require capacity building and practical information on modern rice production practices. Based on the currently available data we propose the development of interactive decision guides for effective implementation of improved nutrient, water, weed management and varieties as illustrated in Fig. 1. The objective of such decision guides can be twofold (1) to assist farmers in their management decisions using a set of logical steps that take into account the prevailing circumstances at their farm and (2) for researchers, extensionists and farmer to evaluate the interaction of factors important to improve rice production through GAP. IRRI and AfricaRice are in the process of developing such interactive guides, to be used on mobile devices e.g. smart-phones and electronic tablets. Soils, broadly described by colour and texture as light sandy or heavy clays and positions on the catena, synonymous to upland, hydromorphic and lowland are considered to require distinctly different management. For clayey soils colour and texture classes include red clays and black clays. Haefele et al. (2006) simplified categorization of soils into (1) inherently poor soils, (2) soils with abnormally poor fertilizer response, (3) soils productive only when fertilizer is applied and (4) soils with normal to high fertilizer response, an approach which is very useful in developing a decision guide for soil fertility management. A soil fertility management decision tool, called Nutrient Manager, is currently in an advanced stage of development for use

52

N. Nhamo et al. / Agricultural Systems 131 (2014) 45–55

Table 4 Summary principles of the two main rice systems applicable to ESA described in literature (a) Integrated Crop Management (ICM) and (b) the System of Rice Intensification (SRI) and Conservation Agriculture (CA), and suggested areas of improvements. Rice system

Principles

Areas of improvement

Sources

Integrated Crop Management (ICM)

(1) Rice cultivation is a production systems involving several components from land preparation to harvest and postharvest management (2) ICM takes into consideration all factors that impact crop growth, yield, quality and profitability

a. Developing effective communication means for farmers

Donovan et al. (1999); Haefele et al., 2000, 2001, 2003; Kebbeh and Miezan, 2003, Wopereis et al., 1999

b. Incorporating local adaptations through farmer experimentation. Improving links to markets by using a value chain approach

(3) Technologies are delivered as a basket of integrated management options (4) Farmers are heterogeneous and different in terms of access to resources (5) Components of ICM include improved varieties, good quality planting materials, improved fertilizers, weed and water management, and efficient and cost effective harvest and post-harvest technologies (6) The cropping calendar is made up of 5 different periods (a) land preparation, basal fertilizer application and sowing (b) first urea application and chemical or manual weeding (c) second urea application and additional weeding (d) final urea application and additional weeding and (e) harvest and post-harvest operations. In addition, N fertilizer is applied at 250–350 kg ha1, water level for each fertilizer application is 3 cm high and weeding methods can be chemical manual or mechanical, but need proper timing System of Rice Intensification (SRI)

(1) Shallow transplanting (1–2 cm) of young seedlings (<16 day-old) into moist but not flooded seed bed

a. Reducing labour requirements on weeding

(2) transplanting of single seedlings per hill

b. Develop local adaptations to recommendations of organic matter inputs

(3) wide plant spacing (25 cm  25 cm to 30 cm  30 cm) (4) alternate wetting-and-drying water regimes during vegetative growth (5) early and regular weeding using a rotary weeder (6) high nutrient inputs in organic form e.g. compost

Conservation Agriculture (CA)

c. Include responses of improved cultivars and synergies from other factors d. Improving links to markets (using value chain approach) e. Adapting water management to local irrigation infrastructure or organization

(1) Use of minimum tillage

a. Finding suitable cover crops and grain legumes.

(2) Soil protection through mulching, dead or living organic materials

b. Reducing labour and/or costs of weed and organic matter management c. Developing suitable tillage equipment d. Creating better links to markets (using value chain approach) e. Improving suitability for crop-livestock farming systems

(3) Use of crop rotation or intercropping (crop combinations) (4) Biomass production whenever possible throughout the whole year (5) Use of multifunctional cover crops

in SSA. Attributes on soil fertility, depth, structure and colour and use of indicator plants, texture, consistence and parent material have successfully been combined by farmers and researchers in arriving at different soil types and associated best management

Deb et al. (2012), Krupnik et al. (2012a–c), Menete et al. (2008), Moser and Barret (2003), Noltze et al. (2012), Stoop et al. (2002), Styger et al. (2011), Tsujimoto et al. (2009)

Dobermann (2004), Erenstein (2011), Giller et al. (2009), Oicha et al. (2010), Scopel et al. (2012)

options (Habarurema and Steiner, 1997). However, lack of correlation between local knowledge of soils and scientifically determined values challenges farmer participatory soil suitability evaluations (Birmingham, 2013). For effective weed management,

N. Nhamo et al. / Agricultural Systems 131 (2014) 45–55

identification of dominant or persistent weed species or groups is an important first step. A proper diagnosis of the weed problem forms the basis for effective weed management recommendations. A decision support tool (Weed Manager) is currently in an early stage of development and will include a recently finalized weed identification tool and data base for weeds of rice in Africa, called AFROweeds, freely accessible online and offline as application on tablets, smartphones and computers (CIRAD-AfricaRice, 2012). There is a high probability of developing effective decision guides with rice farmers especially because of the high market value of rice and the current drive in improving productivity. However, it is important to involve farmers in the development of such tools as this gives ownership to users, ensures thorough testing and understanding on how they lead to efficient resource use management.

5. Implications for research Yield targets that incorporate farmers’ socio-economic status, field soil conditions, available improved technologies and rice cultivars need to be worked out together with farmers in order to reduce yield gaps in rice production systems. Banwo and Makundi (2002) identified promotion of integrated crop management practices, deployment of new technologies, reduction of post-harvest losses and characterization of the nature of gaps for different farm conditions (location specific) as key requirements in narrowing yield gaps. Yield gains from improved weed control and N management have been reported by Becker et al. (2003). When combined with effective water management targeting weeds and soil fertility can lead to significant rice yield increases. Similarly, exploring linkages between biophysical and socio-economic factors can unlock production potential at minimal costs to the farmers (Stoop et al., 2002). Combining a strategic with an adaptive approach (on-farm with full farmer participation) has also been emphasized (Giller et al., 2011). Using a similar approach, the gains derived from use of improved fertilizer, water and weed management on improved cultivars (e.g., Table 3) can be further characterized taking into consideration the impacts of socio-economic and climate variations. Based on our analysis, the following issues are (not exclusively) considered important for near-future research and development efforts: (1) Flexible fertilizer application strategies – that take into account the target yields (short term) and minimization of environmental damage from over application – need to be worked out together with farmers and other stakeholders. Site-specific nutrient management practices need to replace the blanket fertilizer recommendations approach. (2) How to tailor technologies to specific farm conditions need to be investigated in order to increase the implementation of locally adapted Good Agricultural Practices. The cost of conducting these studies can be reduced by delineating areas with similar ecological characteristics, i.e. farm typologies according to Giller et al. (2011), and by integrating technical, farmer management and advisory services. This should lead to best-fit technologies for smallholder farmers, i.e. improved technologies tailored to farmers’ production conditions (e.g., Giller et al., 2011; Le Gal et al., 2011). (3) Locally adapted Good Agricultural Practices need to be developed aimed at intensification and diversification of rice-based production systems, with (a large number of) farmers and integrated in rice value chains, improving the link between (market) demand and supply as well as quality of the produce.

53

(4) Impact of climate variability (extremes) on rice systems need to be investigated through a combination of experiments, farmer recall surveys and crop modelling. Suitable, affordable and effective technologies that contribute to mitigating, or coping with, climate irregularities or other environmental stresses imposed by climate change, need to be developed or adapted with farmers. (5) Crop combinations and sequences that effectively utilize temporal and spatial opportunities for improving rice systems need to be developed and promoted. Introduction of effective legumes and high value crops in rice systems to generate incomes for farmers and effectively utilize water and nutrient resources need to be considered for increase production, productivity and profitability. (6) Locally improved rice systems using relevant elements of potential crop production systems (e.g. SRI, IRM, CA) need to be developed with farmers and other relevant stakeholders. (7) In order to provide incentives associated with adoption of Good Agricultural Practices there is need to evaluate the effectiveness of the value chain approach on crop such as rice in ESA. More importantly, the knowledge gaps in the mechanism associated with adoption of integrated technologies by farmers need to be addressed. 6. Conclusions In East and Southern Africa, rice yield gains following use of improved cultivars and component technologies improving nutrient, water and weed management are evident. Although based on few data, we found that improving weed management could result in the highest relative yield gains (91.6%). There is scope to obtain higher yield gains if technologies are integrated as Good Agricultural Practices for rice however, such gains emanating from locally adapted technology combinations are yet to be thoroughly studied. Yield gaps explained by the socio-economic circumstances of farmers remain a major challenge because rice farmers in East and Southern Africa are often resource constrained and do not have the organizational skills and know-how required for timely and proper implementation of improved technologies. Fertilizer use on rice can be profitable to farmers but the price of such inputs is the most critical factor. Further, there is a paucity of data at present to fully characterize the technical yield gaps in a step by step manner in rice systems. There is however evidence that locally adapting rice systems i.e., with elements of ICM, SRI or CA, can provide farmers with adequate solutions to stagnant rice yields. We propose the further development and implementation of interactive decision support guides for smallholder farmers, focussing on Good Agricultural Practices and based on farm typologies, determining the local availability of resources, for rice-based production systems. Acknowledgements The authors are grateful to two reviewers who provided useful comments for improvement of this paper. References Abo, M.E., Sy, A.A., Alegbego, M.D., 1998. Rice Yellow Mottle Virus (RYMV) in Africa: Evolution, distribution, economic significance on sustainable rice production and management strategies. J. Sustain. Agr. 11, 85–111. Acres, B.D., Rains, A.B., King, R.B., Lawton, R.M., Mitchell, A.J.B., Rackham, L.J., 1985. African dambos: their distribution, characteristics and use. Zeitschrift Fur Geomorphologie, 63–86. Akanvou, R., Kropff, M.J., Bastiaans, L., Becker, M., 2002. Evaluating the use of two contrasting legume species as relay intercrop in upland rice cropping systems. Field Crops Res. 74, 23–36.

54

N. Nhamo et al. / Agricultural Systems 131 (2014) 45–55

Alemu, G., Bayu, W., 2005. Effects if farmyard manure and combined N and P fertilizer on sorghum and soil characteristics in Northeastern Ethiopia. J. Sustain. Agr. 26, 23–41. Ampong-Nyarko, K., De Datta, S.K., 1991. A Handbook for Weed Control in Rice International Rice Research Institute, Manila. Andriesse, W., Fresco, L.O., Duivenbooden, N., Windmeijer, P.N., 1994. Multi-scale characterization of inland valley agro-ecosystems in West Africa. Neth. J. Agr. Sci. 42, 159–179. Balasubramanian, V., Sié, M., Hijmans, R.J., Otsuka, K., 2007. Increasing rice production in Sub-Saharan Africa: challenges and opportunities. Adv. Agron. 94, 55–133. Banwo, O.O., Makundi, R.H., 2002. Bionomics and importance of two species of Chaetocnema in Rice Yellow Mottle Virus Transmission in lowland rice in Tanzania. Phytoparasitica 30, 96–104. Bationo, A., Hartemink, A., Lungu, O., Nai, M., Oko, P., Smali, E., Thiombiano, L., 2006 African Soils: Their Productivity and Profitability of Fertilizer Use. Africa Fertilizer Summit: Nourish the Soil, Feed the Continent, Abuja. Becker, M., Johnson, D.E., 1999. The role of legume fallows in intensified upland rice-based systems of West Africa. Nut. Cyc. Agroecosys. 53, 71–81. Becker, M., Johnson, D.E., 2001. Improved water control and crop management effects on lowland rice productivity in West Africa. Nut. Cyc. Agroecosys. 59, 119–127. Becker, M., Ladha, J.K., Ali, M., 1995. Gren manure technology: potential, usage and limitations. A case study for lowland rice. Plant Soil. 174, 181–195. Becker, M., Johnson, D.E., Wopereis, M.C.S., Sow, A., 2003. Rice yield gaps in irrigated systems along an agro-ecological gradient in West Africa. J. Plant Nutr. Soil Sci. 166, 61–67. Bekunda, M.A., Woomer, P.L., 1996. Organic resource management in banana based cropping systems of the Lake Victoria basin, Uganda. Agr. Ecosyst. Environ. 59, 171–180. Bekunda, M.A., Nkonya, E., Mugendi, D.N., Msaky, J.J., 2002. Soil fertility status, management and research in east Africa. East African J. Rural Dev., 1–24. Birmingham, D.M., 2013. Local knowledge of soils: the case of contrast in Cote d’Ivoire. Geoderma 111, 481–502. Brar, D., Virk, P., 2010. How a modern rice variety is bred. Rice Today 1, 11–12. CIRAD-AfricaRice, 2012. AFROweeds. Montpellier, France-Cotonou, Benin. Cissoko, M., Boisnard, A., Rodenburg, J., Press, M.C., Scholes, J.D., 2011. New Rice for Africa (NERICA) cultivars exhibit different levels of post-attachment resistance against the parasitic weeds Striga hermonthica and Striga asiatica. New Phytol. 192, 952–963. De Wit, C.T., 1992. Resource use efficiency in agriculture. Agri. Syst. 40, 125–151. Deb, D., Lässig, J., Kloft, M., 2012. A critical assessment of the importance of seedling age in the System of Rice Intensification (SRI) in eastern India. Expl. Agric. 48 (3), 326–346. Demont, M., Rizzotto, A., 2012. Policy sequencing and the development of rice value chains in Senegal. Dev. Policy Rev. 30, 451–472. Demont, M., Jouve, P., Stessens, J., Tollens, E., 2007. Boserup versus Malthus revisited: evolution of farming systems in northern Cote d’Ivoire. Agri. Syst. 93, 215–228. Dobermann, A., 2004. A critical assessment of the System Of Rice Intensification (SRI). Agri. Syst. 79, 261–281. Donovan, C., Wopereis, M.C.S., Guindo, D., Nebie, B., 1999. Soil fertility management in irrigated rice systems in the Sahel and Savanna regions of West Africa. Part II. Profitability and risk analysis. Field Crops Res. 61, 147–162. Ducrot, R., Capillon, A., 2004. A practice analysis to account for adoption of innovations in irrigated rice cropping system in Lake Alaotra (Madagascar). J. Sustain. Agr. 24, 71–96. Erenstein, O., 2011. Cropping systems and crop residue management in the TransGangetic Plains: issues and challenges for conservation agriculture from village surveys. Agri. Syst. 104, 54–62. Esilaba, A.E., Nyende, P., Nalukenge, G., Byalebeka, J.B., Delve, R.J., Ssali, H., 2005. Resource flows and nutrient balances for crop and animal production in smallholder farming systems in eastern Uganda. Agr. Ecosyst. Environ. 109, 192–201. FAO, 2010. FAO Statistical Databases – Agriculture. . Garenne, M., 2002. The political economy of an urban famine. IDS Bullet. 33, 55–62. Giannin, A., Biasutti, M., Held, I.M., Sobel, A.H., 2008. A global perspective of African climate. Clim. Change 90, 359–383. Giller, K.E., 2001. Nitrogen Fixation in Tropical Cropping Systems, second ed. CABI International, Wallingford. Giller, K.E., Witter, E., Corbeels, M., Tittonell, P., 2009. Conservation agriculture and smallholder farming in Africa: the heretics’ view. Field Crops Res. 114, 23– 34. Giller, K.E., Tittonell, P., Rufino, M.C., van Wijk, M.T., Zingore, S., Mapfumo, P., AdjeiNsiah, S., Herrero, M., Chikowo, R., Corbeels, M., Rowe, E.C., Baijukya, F., Mwijage, A., Smith, J., Yeboah, E., van der Burg, W.J., Sanogo, O.M., Misiko, M., de Ridder, N., Karanja, S., Kaizzi, C., K’ungu, J., Mwale, M., Nwaga, D., Pacini, C., Vanlauwe, B., 2011. Communicating complexity: Integrated assessment of trade-offs concerning soil fertility management within African farming systems to support innovation and development. Agri. Syst. 104, 191–203. Habarurema, E., Steiner, K.G., 1997. Soil suitability classification by farmers in southern Rwanda. Geoderma 75, 75–87. Haefele, S.M., Johnson, D.E., Diallo, S., Wopereis, M.C.S., Janin, I., 2000. Improved soil fertility and weed management is profitable for irrigated rice farmers in Sahelian West Africa. Field Crops Res. 66, 101–113.

Haefele, S.M., Wopereis, M.C.S., Donovan, C., Maubuisson, J., 2001. Improving the productivity and profitability of irrigated rice production in Mauritania. Eur. J. Agron. 14, 181–196. Haefele, S.M., Wopereis, M.C.S., Ndiaye, M.K., Kropff, M.J., 2003. A framework to improve fertilizer recommendations for irrigated rice in West Africa. Agri. Syst. 76 (1), 313–335. Haefele, S.M., Naklang, K., Harnpichitvitaya, D., Jearakongman, S., Skulkhu, E., Romyen, P., Tabtim, S., Suriya-arunroj, D., Khunthasuvon, S., Kraisorakul, D., Youngsuk, P., Amarate, S.T., Wale, L.J., 2006. Factors affecting rice yield and fertilizer response in rainfed lowlands of northeast Thailand. Field Crops Res. 98, 39–51. Hartemink, A.E., 1997. Soil fertility decline in some major soil groupings under permanent cropping in Tanga region, Tanzania. Geoderma 75, 215–229. Hartemink, A.E., Priess, J., Veldkamp, E., Teketay, D., Lesschen, J.P., 2005. Assessment of soil nutrient depletion and its spacial variability on smallholders’ mixed farming systems in Ethiopia using partial versus full nutrient balances. Agr. Ecosyst. Environ. 108, 1–16. Jagadish, S.V.K., Septiningsih, E.M., Kohli, A., Homson, M.J., Ye, C., Redona, E., Kumar, A., Gregorio, G.B., Assmann, R., Ismail, A.M., Singh, R.K., 2012. Genetic advances in adapting rice to a rapidly changing climate. J. Agron. Crop Sci. 198, 360–373. Jamil, M., Rodenburg, J., Charnikhova, T., Bouwmeester, H.J., 2011. Pre-attachment Striga hermonthica resistance of NERICA cultivars based on low strigolactone production. New Phytol. 192, 964–975. Kaihura, F.B., Kullaya, I., Kilasara, M., Lal, R., Singh, B.R., Aune, J.B., 1996. Topsoil thickness effects on soil properties and maize (Zea mays) yield in three ecoregions of Tanzania. J. Sustain. Agr 9, 11–30. Kaihura, F.B.S., Kullaya, I.K., Kilasara, M., Aune, J.B., Singh, B.R., Lal, R., Arshad, M.A., 1999. Soil quality effects of accelerated erosion and management systems in three eco-regions of Tanzania. Soil Till. Res. 53, 59–70. Kajiru, G.J., Kileo, R.O., Stroud, A., Budelman, A., 1998. Validating a fertilizer recommendation across a diverse cropping environment. Nut. Cyc. Agroecosys. 51, 163–173. Kajisa, K., Payongayong, E., 2011. Potential of and constraints to the rice Green Revolution in Mozambique: a case study of the Chokwe irrigation scheme. Food Policy 36, 615–626. Kanyeka, Z.K., Msomba, S.W., Kihupi, A.N., Penza, M.S.F., 1996. Rice ecosystems in Tanzania: characterization and classification. Res. Train. Newsl. 9, 13–15. Kayeke, J., Siguba, P.K., Msaky, J.J., Mbwaga, A., 2007. Green manure and inorganic fertilizer as management strategies for witchweed and upland rice. Afr. Crop Sci. J. 15, 161–171. Kebbeh, M., Miezan, K.M., 2003. Ex-ante evaluation of integrated crop management options for irrigated rice production in the Senegal River Valley. Field Crops Res. 81, 87–94. Kennedy, E., Reardon, T., 1994. Shift to non-traditional grain in the diets of east and west Africa: role of women’s opportunity cost of time. Food Policy 19, 45–56. Kijima, Y., Otsuka, K., Sserunkuuma, D., 2011. An inquiry into constraints on a Green Revolution in Sub-Saharan Africa: the case of NERICA rice in Uganda. World Dev. 39, 77–86. Kijima, Y., Sserunkuuma, D., Otsuka, K., 2006. How revolutionary is the ‘‘NERICA Revolution’’? Evidence from Uganda. The Developing Economies XLIV (2), 252– 267. Krupnik, T.J., Shennan, C., Settle, W.H., Demont, M., Ndiaye, A.B., Rodenburg, J., 2012a. Improving irrigated rice production in the Senegal River Valley through experimental learning and innovation. Agri. Syst. 109, 101–112. Krupnik, T.J., Rodenburg, J., Haden, V.R., Mbaye, D., Shennan, C., 2012b. Genotypic trade-offs between water productivity and weed competition under the System of Rice Intensification in the Sahel. Agric. Water Manage. 115, 156–166. Krupnik, T.J., Shennan, C., Rodenburg, J., 2012c. Yield, water productivity and nutrient balances under the System of Rice Intensification and recommended management practices in the sahel. Field Crops Res. 130, 155–167. Lawrence, P.R., Dijkman, J.T., Jansen, H.G.P., 1997. The introduction of animal traction into inland valley regions. 1. Manual labour and animal traction in the cultivation of rice and maize: a comparison. J. Agric. Sci. 129, 65–70. Le Gal, P.-Y., Dugue, P., Faure, G., Novak, S., 2011. How does research address the design of innovative agricultural production systems at farm level? A review. Agri. Syst. 104, 714–728. Lobell, D.B., Cassman, K.G., Field, C.B., 2009. Crop yield gaps: their importance, magnitudes and causes. Ann. Rev. Environ. Resources 34, 179–204. Loevinsohn, M.E., Mugarura, J., Nkusi, A., 1994. Cooperation and innovation by farmer groups: scale in the development of Rwandan Valley farming systems. Agri. Syst. 46, 141–155. Luzi-Kihupi, A., Tamura, M., Mbapila, J.C., 2004. Rice improvement and production in Tanzania. JICA Seminar on Promotion and Dissemination in Africa AIAD Japan International Cooperation Agency, Nairobi, pp. 55–90. Luzi-Kihupi, A., Mlozi, M.R.S., Nchimbi-Msolla, S., 2007. Cooking and eating quality of rice yellow mottle virus resistant rice mutants: its implications for the future breeding work. Tanzania J. Agric. Sci. 8, 193–202. Luzi-Kihupi, A., Zakayo, J.A., Tusekelege, H., Mkuya, M., Kibanda, N.J.M., Khatib, K.J., Kaerere, A., 2009. Mutation breeding for rice improvement in Tanzania. In: Shu, Q.Y. (Ed.), Induced Plant Mutations in the Genomics Era Food and Agriculture Organization of the United Nations, Rome, pp. 385–387. Mafongoya, P.L., Bationo, A., Kihara, J., Waswa, B.S., 2006. Appropriate technologies to replenish soil fertility in Southern Africa. Nut. Cyc. Agroecosys. 76, 137–151. Makoye, J.J., Winge, N.J., 1996. Field activities in Shinyanga Region 1994/95. Kilimo/ FAO Plant Nutrition Programme, URT/FAO, Dar es Salaam.

N. Nhamo et al. / Agricultural Systems 131 (2014) 45–55 McCartney, M.P., Houghton-Carr, H.A., 2009. Working wetlands potential: an index to guide the sustainable development of African wetlands. Nat. Resources Forum 33, 99–110. Meertens, H.C.C., Roling, N.G., 2000. Non-adoption of rice fertilizer technology based on a farming systems research-extension methodology in Sukumaland, Tanzania: a search for reasons. J. Extension Syst. 16, 1–22. Meertens, H.C.C., Ndege, L.J., Lupeja, P.M., 1999. Cultivation of rainfed, lowland rice in Sukumaland, Tanzania. Agr. Ecosyst. Environ. 76, 31–45. Meertens, H.C.C., Kajiru, G.J., Ndege, L.J., Enserink, H.J., Brouwer, J., 2003. Evaluation of on-farm soil fertility research in the rainfed lowland rice fields of Sukumaland, Tanxania. Exp. Agr. 39, 65–79. Menete, M.Y.L., van Es, H.M., Brito, R.M.L., DeGloria, S.D., 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. Mghase, J.J., Shiwachi, H., Nakasone, K., Takahashi, H., 2010. Agronomic and socioeconomic constraints to high yield of upland rice in Tanzania. Afr. J. Agric. Res. 5, 150–158. Moser, C.M., Barrett, C.B., 2003. The disappointing adoption dynamics of a yield increasing, low external-input technology: the case of SRI in Madagascar. Agri. Syst. 76, 1085–1100. Msomba, S.W., Kibanda, J.M., Penza, H., Tusekelege, H., Mkuya, M., Kafiriti, E., Mbapila, J.C., Kanyeka, Z.K., 2004. TXD306 – a high-yielding and mediummaturity aromatic rice for the rainfed lowland and irrigated ecosystems in Tanzania. Int. Rice Res. News 29, 29–30. Mulwafu, W.O., Nkhoma, B.G., 2002. The use and management of water in the Likangala irrigation scheme complex in southern Malawi. Phys. Chem. Earth 27, 839–844. Munene, J.T., Kinyamario, J.I., Holst, N., Mworia, J.K., 2008. Competition between cultivated rice (Oryza sativa) and wild rice (Oryza punctata) in Kenya. Afr. J. Agric. Res. 3, 605–611. Mwaseba, D.L., Kaarhus, R., Johnsen, F.H., Mvena, Z.S.K., Mattee, A.Z., 2006. Beyond adoption/rejection of agricultural innovations: empirical evidence from smallholder rice farmers in Tanzania. Outlook Agri. 35, 263–272. Nakano, Y., Kajisa, K., 2012 The determinants of technology adoption: A case of the rice sector in Tanzania. In: Proceedings of International Association of Agricultural Economists Conference, 18–24 August 2012, Foz do Iguacu, Brazil. Nandwa, S.M., Bekunda, M.A., 1998. Research on nutrient flows and balances in East and Southern Africa: state-of-the-art. Agr. Ecosyst. Environ. 71, 5–18. Nhamo, N., Sithole, T., Dhliwayo, D.K.C., Mupangwa, W., Gatsi, T., Chikazunga, D., Nemasasi, H., 2002. The prospects of smallholder farmers utilizing organic materials as soil fertility amendments in Murewa and Mvuma, Zimbabwe. A Soil Fertility Network Research Results Working Paper No. 9 International Maize and Wheat Improvement Centre (CIMMYT), Harare. Nkonya, E., Kaizzi, C., Pender, J., 2005. Determinant of nutrient balances in a maize farming system in eastern Uganda. Agri. Syst. 85, 155–182. 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. Agri. Syst. 108, 64–73. NPA, 2007. Marketing of Rice based on Building a Strong Value-Chain for the Rice Industry in Uganda. National Planning Authority, Kampala, pp. 1–93. Oicha, T., Cornelis, W.M., Verplancke, H., Nyssen, J., Govaerts, B., Behailu, M., Haile, M., Deckers, J., 2010. Short-term effects of conservation agriculture on Vertisols under tef (Eragrostis tef (Zucc.) Trotter) in the northern Ethiopian highlands. Soil Till. Res. 106, 294–302. Okalebo, J.R., Othieno, C.O., Woomer, P.L., Karanja, N.K., Semoka, J.R.M., Bekunda, M.A., Mugendi, D.N., Muasya, R.M., Bationo, A., Mukhwana, E.J., 2006. Available technologies to replenish soil fertility in East Africa. Nut. Cyc. Agroecosys. 76, 153–170. Otsuka, K., Kalirajan, K.P., 2005. An overview and prospects for a green revolution in Sub-Saharan Africa. J. Agric. Dev. Econ. 2, 1–6. Otsyina, R., Asenga, D., Mumba, M., 1995. Annual Progress Report 1995. Tanzania/ ICRAF Agroforestry Research Project Shinyanga, Shinyanga. Perret, S.R., Stevens, J.B., 2006. Socio-economic reasons for the low adoption of water conservation technologies by smallholder farmers in Southern Africa: a review of the literature. Dev. Southern Africa 23, 461–476. Piha, M.I., 1993. Optimizing fertilizer use and practical rainfall capture in a semiarid environment with variable rainfall. Exp. Agr. 29, 405–415. Poulton, C., Kydd, J., Dorward, A., 2006. Increasing Fertilizer use in Africa: What have we Learned? Agriculture and Rural Development Discussion Paper 25. The World Bank, Washington. Rabbinge, R., 1993. The Ecological Background to Food Production. John Wiley & Sons, Chichester, pp. 2–29. Raes, D., Kafiriti, E.M., Wellens, J., Deckers, J., Maertens, A., Mugogo, S., Dondeyne, S., Descheemaeker, K., 2007. Can soil unds increase the production of rain-fed lowland rice in south eastern Tanzania. Agri. Water Manage. 89, 229–235. Renkow, M., Byerlee, D., 2010. The impacts of CGIAR research: a review of recent evidence. Food Policy 35, 391–402.

55

Riches, C.R., Mbwaga, A.M., Mbapila, J., Ahmed, G.J.U., 2005. Improved weed management delivers increased productivity and farm incomes from rice in Bangladesh and Tanzania. Asp. Appl. Biol. 75, 127–138. Rodenburg, J., Demont, M., 2009. Potential of herbicide-resistant rice technologies for Sub-Saharan Africa. AgBioForum 12, 313–325. Rodenburg, J., Johnson, D.E., 2009. Weed management in rice-based cropping systems in Africa. Adv. Agron. 103, 149–218. Rodenburg, J., Saito, K., Kakai, R.G., Toure, A., 2009. Weed competitiveness of the lowland rice varieties of NERICA in the southern Guinea Savanna. Field Crops Res. 114, 411–418. Rodenburg, J., Riches, C.R., Kayeke, J.M., 2010. Addressing current and future problems of parasitic weeds in rice. Crop Prot. 92, 210–221. Rodenburg, J., Meinke, H., Johnson, D.E., 2011. Challenges for weed management in African rice systems in a changing climate. J. Agric. Sci. 149, 427–435. Rodenburg, J., Zwart, S.J., Kiepe, P., Narteh, L.T., Dogbe, W., Wopereis, M.C.S., 2013. Sustainable rice production in African inland valleys: seizing regional potentials through local approaches. Agri. Syst., in preparation. Rugumamu, C.P., 2014. Empowering smallholder rice farmers in Tanzania to increase productivity for promoting food security in Eastern and Southern Africa. Agric. Food Security 3 (7), 1–8. Sakane, N., Alvarez, M., Becker, M., Bohme, B., Handa, C., Kamiri, H.W., Langensiepen, M., Menz, G., Misana, S., Mogha, N.G., Moseler, B.M., Mwita, E.J., Oyieke, H.A., van Wijk, M.T., 2011. Classification, characterization and use of small wetlands in East Africa. Wetlands 31, 1103–1116. Scopel, E., Triomphe, B., Affholder, F., Da Silva, F.A.M., Corbeels, M., Xavier, J.H.V., Lahmar, R., Recous, S., Bernoux, M., Blanchart, E., Mendes, I.C., De Tourdonnet, S., 2012. Conservation agriculture cropping systems in temperate and tropical conditions, performances and impacts. A review. Agron. Sustain. Dev. http:// dx.doi.org/10.1007/s13593-012-0106-9. Seck, P.A., Tollens, E., Wopereis, M.C.S., Diagne, A., Bamba, I., 2010. Rising trends and variability of rice prices: threats and opportunities for Sub-Saharan Africa. Food Policy 355, 403–411. Seck, P.A., Diagne, A., Mohanty, S., Wopereis, M.C.S., 2012. Crops that feed the world: rice. Food Policy 4, 7–24. Shayo, B.H., Mamiro, P., Nyaruhucha, C.N.M., Mambolea, T., 2006. Physico-chemical characteristics of selected rice cultivars grown in Morogoro. Tanz. J. Sci. 32 (1), 29–35. Singh, A.K., 2006. Exploring water nutrient synergies – a must for evergreen revolution. J. Indian Soc. Soil Sci. 54, 396–407. Sistani, K.R., Reddy, K.C., Kanyika, W., Savant, N.K., 1998. Integration of rice crop residue into sustainable rice production system. J. Plant Nutr. 21, 1855– 1866. Sseguya, H., Semana, A.R., Bekunda, M.A., 1999. Soil fertility management in the banana based agriculture of central Uganda: farmers constraints and opinions. Afr. Crop Sci. J. 7, 559–567. Stoop, W.A., Uphoff, N., Kassam, A., 2002. A review of agricultural research issues raised by the System of Rice Intensification (SRI) from Madagascar: opportunities of improving farming systems for resource poor farmers. Agri. Syst. 71, 249–274. Stoop, W.A., Adam, A., Kassam, A., 2009. Comparing rice production systems: a challenge for agronomic research and for the dissemination of knowledgeintensive farming practices. Agri. Water Manage. 96, 1491–1501. Styger, E., Ag Atter, M., Guindo, H., Ibriham, 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. Toure, A., Becker, M., Johnson, D.E., Kone, B., Kossou, D.K., Kiepe, P., 2009. Response of lowland rice to agronomic management under different hydrological regimes in an inland valley of Ivory Cost. Field Crops Res. 114, 304–310. Tran, D., 2004. Rice and Narrowing the Yield Gap. Food and Agriculture Organization of the United Nations (FAO), Rome. Tsujimoto, Y., Horie, T., Randriamihary, H., Shiraiwa, T., Homma, K., 2009. Soil management: the key factor for higher productivity in the fields utilizing the System of Rice Intensification SRI in the central highlands of Madagascar. Agri. Syst. 100, 61–71. van der Heyden, C.J., New, M.G., 2003. The role of a dambo in the hydrology of a catchment and the river network downstream. Hydrol. Earth Syst. Sc. 7, 339– 357. Verhoeven, J.T.A., Setter, T.L., 2010. Agricultural use of wetlands: opportunities and limitations. Annu. Bot. 105, 155–163. Wopereis, M.C.S., Donovan, C., Nebie, B., Guindo, D., N’-Diaye, M.K., 1999. Soil fertility management in irrigated rice systems in the Sahel and Savanna regions of West Africa. Part I. Agronomic analysis. Field Crops. Res., 125–145. Wopereis, M.C.S., Diagne, A., Rodenburg, J., Sie, M., Somado, E.A., 2008. Whu NERICA is a successful innovation for African farmers: a response to Orr. et al. from Africa Rice Center. Outlook Agri. 37, 169–176. Zimdhal, R.L., 2007. Fundamentals of Weed Science. Elsevier Inc., London.