Weed Management in Rainfed Upland Rice Fields under Varied Agro-Ecologies in Nigeria

Oyebanji O. Alagbo, Oluyemisi A. Akinyemiju, Bhagirath S. Chauhan

Review

Weed Management in Rainfed Upland Rice Fields under Varied Agro-Ecologies in Nigeria

Oyebanji O. Alagbo1, Oluyemisi A. Akinyemiju2, Bhagirath S. Chauhan3

(Department of Crop, Soil and Pest Management, Ondo State University of Science and Technology, Okitipupa 353, Ondo-State, NigeriaQueensland Alliance for Agriculture and Food Innovation)

The demand for rice to meet the dietary need in low-income countries is expected to witness an exponential rise as the population increases. Meeting the rice demand domestically has remained challenging due to significant yield loss caused by several biotic and abiotic factors. Among these factors, one of the most important is the high weed pressure that ravages the upland rice ecology. In Nigeria, several independent weed control techniques, such as physical, chemical and cultural methods, have been recommended and adopted for weed control across varying rice upland ecologies. However, outcomes of these approaches when used independently have not consistently led to an increase in yield. There remains an outstanding deficit between the actual yield and the potential rice yield. This review aimed to identify potential research gaps, and quest effective and sustainable weed management strategies in smallholder upland rice farming systems in Nigeria. A critical analysis of studies suggests the potential of sustainable weed management practices if adopted and adapted smartly in different upland ecologies in Nigeria. Competitiveness of upland rice against weeds can be enhanced through strategic integration of weed competitive cultivars, optimum nitrogen application timings (within weed-free periods), uniform plant spacing, and high seeding rates, with conventional herbicide/manual weed control practices.However, such management practices can only be engaged where inputs are supplied on time and the technical know-how is extended to farmers. The review equally highlights potential research gaps for further studies.

herbicide; integrated weed management; weed flora; tillage practice; rainfed upland rice

Rice is one of the predominant staple foods in low- income countries and a major source of carbohydrate. It also contains protein, fat, thiamine, riboflavin, zinc and niacin in minimal amounts (Phillip et al, 2018). Cultivated upland rice area is expected to increase in the future due to shortages in labor and water supply globally (Mahajan et al, 2014). To meet the current food security challenge in sub-Saharan Africa, cultivatedupland rice area has increased in recent years (Saito et al, 2018). Between 0.5–1.0 million hectares of cultivated area in Nigeria contributes 20%–30% of the global annual upland rice production (Saito et al, 2018).

The upland rice ecology covers 32%–55% of the total land area used for rice cultivation in Nigeria (Biyi, 2005; Phillip et al, 2018). Upland rice thrives well on heavy soils with good water holding capacity, a slightly acidic to neutral pH (5.5–7.0), and an average rainfall of 1 200–1 600 mm (Adeosun et al, 1996; Kamai et al, 2020). Nevertheless, achieving optimum yield in the sub-Saharan African region is still difficult due to the high incidence of insect pests, birds, diseases, weeds anddrought, as well as limited technical know-how among smallholder farmers (Akobundu 1987; Rodenburg and Johnson, 2009; Nianget al, 2017; Kamai et al, 2020). For instance, in Nigeria, upland rice yields in farmers’ fields rarely exceed 1–2 t/hm2(Akobundu, 1987; Phillip et al, 2018),compared with an estimated yield of 4–6 t/hm2under experimental field conditions (Akobundu, 1987; Singh et al, 1997; Adeosun and Lagoke, 2005; Niang et al, 2017; Kamai et al, 2020).

Amongst these challenges, the increasing pressure of weeds in the upland rice ecology has remained a major cause of rice yield loss in Nigeria (Akobundu, 1987; Emmanuel et al, 2021b). More importantly, the increased abundance and diversity of weed populations in different upland agro-ecologies calls for concern. Uncontrolled weed infestation in upland rice accounts for 42%–100% yield reductions (Akobundu, 1987; Ekeleme et al, 2007; Rodenburg and Johnson, 2009; Daramola et al, 2020; Emmanuel et al, 2021a). To reduce such huge loss, several studies have proposed weed management strategies for upland and lowland rice in sub-Saharan Africa (Rodenburg and Johnson, 2009; Saito et al, 2010; Ogwuike et al, 2014; Johnson et al, 2019; Rodenburg et al, 2019). However weed population and species vary from year to year (Akobundu, 1981; Kolo et al, 2021) and from one location to another, depending on rainfall pattern, species composition, soil fertility level, frequency of land use, seeding technique and weed management strategy under specific agroecology (Akobundu, 1987; Kolo et al, 2021). Therefore, managing weeds efficiently under changing climate would further necessitate better knowledge of weeds peculiar to different rice ecologies and the need to propose specific weed management strategies. It is not clear whether studies conducted so far in Nigeria have some merits to propose sustainable weed management in upland environments. Hence, it is need to review weed management research conducted in upland rice fields under different agro-ecologies in Nigeria. The objectives of this review were to identify potential research gaps, and quest effective and sustainable weed management strategies in smallholder upland rice farming systems in Nigeria.

Weed flora of Nigerian upland rice ecology

Table 1. Predominant weed flora in rainfed upland rice ecologies in Nigeria.

SGS, Southern Guinea Savanna; SSR, South-South Rainforest; SWR, Southwest Rainforest; SER, Southeast Rainforest; NGS, Northern Guinea Savanna; +, Presence of weeds; –, Absence of weeds.

Pre-dominant weed flora of rainfed upland rice and their distributions across different agro-ecologies in Nigeria are shown in Table 1. About 69% of these weeds are annual grasses and broadleaved weeds, some of which have been declared a threat to upland rice ecologies in sub-Saharan Africa and many parts of the world, including:,,,,, and(Rodenburg et al, 2015; Kraehmer et al, 2016). The invasiveness of weeds under any cropping systems has been attributed to several factors. For instance, the prevalence of annual weeds in the upland rice ecology could have resulted from the persistent and repeated deposit of weed seeds in topsoil layers, especially under conservation agriculture systems (Chauhan and Johnson, 2008a, b, c), with poor spacing and seeding rates which are common to most rice ecologies in Nigeria. However, recent weed management studies rarely address the biology and ecology of problem weeds in upland rice fields in Nigeria.

Studies on weed flora dynamics in Nigeria are scanty and often inconsistent, and therefore, it becomes impossible to understand the behaviour of weeds in rice ecologies. A study of weed flora in Edo State of Nigeria (a typical rainforest) was conducted to understand the frequency, abundance and uniformity of weed species across upland rice fields characterized by consistent hoe weeding and/or hand-pulling practices (Obadoni and Remison, 2004). The results are similar to the weed flora reported by Gill and Ene (1978), 26 years earlier in upland rice fields in the same region. Suchsimilarity suggests the use of a consistent agronomic practice that leave the weed flora undisturbed. Unfortunately, recent and increased adoption of herbicide technology in upland rice in Nigeria and the response of weeds to climate change in other parts of the world suggest that the weed community structure across different agroecological zones in Nigeria must have changed with time(Ziska, 2003; Franks et al, 2007; Alagbo, 2016; Adigun et al, 2017). Hence, to model a sustainable weed management strategy, it is necessary to understand how weed species have adapted to agronomic practices in various rice ecologies(Dada et al, 2017a).

Physical methods of weed control in upland rice

Tillage practices

In upland rice, good land preparation combined with timely sowing is known to retard weed infestation significantly (Singh and Ghosh, 1992). Such practice ensures the deep burial of surface weed seeds (Chauhanand Johnson, 2010). Unfortunately, conventional tillage is capital intensive and non-affordable among most rural farming households. About USD 60 is estimated to cultivate 1 hm2field in Nigeria (FAO, 2014), and such a huge cost may be unaffordable by farmers in low-income countries. Besides, timely accessibility to tractors remains a challenge among smallholder farmers in some parts of the country (Ajah, 2014).

Alternatively, zero or minimum tillage is affordable and widely adopted in rainforest zones where rice mechanization is limited due to undulating land terrains and tree vegetations. Nevertheless, such tillage practices have been associated with higher weed pressures in these zones (Olofintoye and Ajayi, 1997; Alagbo, 2016), probably due to the early emergence of weed seeds found within topsoil surfaces with low initial dormancy and high light requirement for germination (Chauhan and Johnson, 2008a, b, c). Unfortunately, the relationship between weed seed germination and tillage practice is largely unknown in Nigerian upland rice ecologies. It is important to study the association between the rising weed pressures and specific tillage practices adopted in specific upland rice ecology.

Practice of stale seedbed technique

The stale seedbed technique is the inducement of weed seed germination after rain or irrigation before planting. Such weeds are then controlled using tillage or herbicides. The purpose is to reduce the population of surface weed seeds before crop seed sowing. The stale seedbed practice is found to reduce the weed population by 53% compared with plots where the technique is not implemented (Singh et al, 2007). In Nigeria, where the practice of conservation tillage is common with prevalent annual weeds in upland rice ecosystems (Table 1), the practice of stale seedbed techniques, combined with active predation of weed seeds by insects, maybe a possible remedy to control vigorous annual weeds.

Mechanical weed control

The use of mechanical weeders has been reported as a good complement for existing weed management practices in sub-Saharan Africa (Rodenburg et al, 2015). Some mechanical weeders were tested recently among selected rice growers in Lafia, Niger State, Nigeria (Johnson et al, 2019),and the farmers prefer the ring hoe mechanical weeder as it reduces weeding time significantly. However, one limitation of this technology is the non-uniformity in row spacing widely practised among upland rice farmers, as the ring hoe requires a uniform spacing,and another concern is whether this technology would be affordable in the long run (Johnson et al, 2019). Hence, it is important to investigate the adoption of this technology in other rice-growing communities in Nigeria.

Manual weeding practices

The traditional weed control method in upland rice involves the use of hoe and/or hand-pulling (Kamai et al, 2020). It is laborious, expensive and limited to small-scale production. In most cases, laborers become scarce at the optimum weeding period due to the tedious nature of manual weeding and demand by other farm activities (Adeosun et al, 1996). Means of 173, 259 and 376 man-hours per hm2are needed to carry out manual weeding once, twice and thrice respectively in sub-Saharan Africa (Ogwuike et al, 2014). Practically, manual weeding often causes some levels of damage to rice due to difficulty in weeding within close inter- and intra-row spacing (Adigun et al, 2005). In addition, weeds with rhizomatous roots and morphology similar to rice often escape manual weeding when rice is sown in narrow spacing (Adigun et al, 2017; Kamai et al, 2020).

Field trials in upland rice often conclude hand- weeding treatments as the best in terms of weed control, good crop vigor and grain yield compared with other treatments in Nigeria (Akobundu, 1987; Adeosun and Lagoke, 2005; Obadoni and Remison, 2007; Adigun et al, 2017). It is practically not possible due to a shortage of labor for weed removal at the critical period, coupled with unfavorable weather conditions (Adigun et al, 2014). In practical terms, manual weeding would be appropriate when integrated with other weed control methods in upland rice.

Cultural weed management practices

The practice of appropriate field operation timing with the use of weed competitive cultivars, fertilizers, narrow crop spacing and high seeding rates to retard weed growth has been proposed several times (Kamara et al, 2010; Kolo et al, 2012, 2021). In most cases, such practices are integrated with a major weed control method. For instance, the use of nitrogen (N) fertilizer, cropping pattern and competitive cultivars are routinely practised in upland rice fields in Nigeria; however, their potentials for integrated weed managementare still unharnessed. This is probably due to poor input supply.

Timeliness of operation

The critical period of weed control largely depends on the relationship between crop seeding date and the emergence period of weed species in specific rice ecology (Yawale et al, 2019). Under direct seeding, the emergence of rice and weed seedlings occurs at the same time (Nojima, 1996), which results in high competition for water, light and other soil nutrients. To attain optimal yields in direct seeding in Nigeria, weeds should be better controlled at the early developmental stages of upland rice (Adigun et al, 2005). Timely and effective weed control enhances nutrient uptake by the rice plants and reduces uptake by weeds (Kolo et al, 2021). Therefore, understanding the critical period of weed removal is a key to develop an effective management strategy in Nigerian upland rice fields. Once weeds are not removed at this period, crop yield loss becomes inevitable. The period between 3 and 9 weeks after sowing is the most critical period of weed competition in upland rice in Nigeria (Adeosun, 2005, 2008; Yawale et al, 2019; Emmanuel et al, 2021a). It is equally agreed that fertilizer application timing should be synchronized with the weed-free period for efficient nutrient uptake by rice (Adeosun, 2005, 2008; Kolo et al, 2021). It is needful that agricultural policies and extension programs provide such an environment whereby smallholder farmers can adhere strictly to such timely practices. Such timeliness is one of the elements of good agronomic practices that can boost upland rice yield in major production zones.

Effect of N on upland rice ecosystem

N remains the most limiting and important soil nutrient in sub-Saharan Africa (Saito et al, 2019). The total N present in Nigerian savanna soils ranges from 0.136%–0.194% (Kamai et al, 2020). Low N levels affect significantly to the yield gap between actual and potential rice yields. Such limited soil N is further competed by rice and weeds (Sankaran and de Datta, 1986). Increased N can enhance dry matter accumulationof weeds competing with upland rice when weeding is delayed (Kolo et al, 2020, 2021), which often results in significant loss of rice yield (Usman et al, 2002; Kolo et al, 2021). The level of N uptake differs among weed species (Kolo et al, 2020). For example, a positive correlation is confirmed between soil N level and N intake by(Ali and Sankaran, 1984) andin upland rice (Chauhan and Johnson, 2010). N can be effectively utilized by rice when the fields are free of weeds (IRRI, 1973).In Nigeria, synchronizing high N application with weed-free periods increases yield and yield components of upland rice cultivars, due to efficient N uptake that enhances rice growth and development under weed-free conditions (Adeosun, 2005, 2008; Kolo et al, 2020, 2021). For instance, early weed removal up to 6 or 9 weeks after sowing combined with N application between 90–120 kg/hm2(depending on rice agro-ecologies) enhances early crop establishment and increases upland rice yield (Usman et al, 2002;Emmanuel et al, 2021b; Kolo et al, 2021). Alternatively, when early weed removal is not feasible, rice competitiveness against weeds at the early growth stage can be achieved with a higher rate of N fertilizer, where weed competitive cultivars are adopted. For example, application of N at 90 kg/hm2on NERICA 8 (an upland variety) significantly increases crop vigor, the number of tillers and plant height at the early growth stage (Kamara et al, 2010; Adigun et al, 2017).

Effects of crop spacing and seeding rate

Maintaining an optimum plant population is necessary to maximize grain yield. Optimum plant density enhances rapid and uniform plant growth in such a way that light, water and nutrient resources are efficiently utilized by crop plants compared with competing weed species (Miah et al, 2004).To enhance early crop establishment and effective competition against weeds, appropriate plant spacing and seeding rate plays a vital role. Inter- and intra-row spacing between 15 and 30 cm can achieve better suppression of weed growth (Olofintoye and Ajayi, 1997; Adeyemi et al, 2015). Also, a significant interaction effect between weeding regime, seeding rate, and grain yield suggests that a proper seeding rate can compliment other weed control methods in upland rice (Kehinde, 2002).

Manual seeding remains a major seeding technique in many upland rice fields in Nigeria. Under such a situation, uniform plant spacing and seeding density, required to maintain optimum plant population, is seldom realised. Thus, achieving better weed suppression is not feasible, even with the best cultivars. In addition, reduced adoption of ring hoe weeding tools in Nasarawa State of Nigeria is recently attributed to the lack of uniform plant spacing practised among rural farmers (Johnson et al, 2019). This further suggests that adopting new weed control technology in Nigeria would necessitate accurate seeding and uniform spacing technique. Hence, advocating the adoption of smart and accurate rice seeders is crucial in upland rice ecologies.

Use of competitive cultivars

Differences in rice varietal genotype, age, morphology and physiology determine, to a large extent, the performance of rice cultivars in the field (Akobundu, 1987). The development of weed-competitive rice cultivars for upland rice production is a key to optimize grain yield in smallholder farms in sub-Saharan Africa (Ekeleme et al, 2009; Saito et al, 2018). Many studies have attributed cultivar-weed competitiveness to a cultivar ability to tolerate or suppress weed growth and maintain high grain yield under both weed-free and weedy conditions(Jannink et al, 2000; Fischer et al, 2001; Zhao et al, 2006; Ekeleme et al, 2009; Saito et al, 2012). Genetic engineering of crop competitive genes against weeds has been proposed (Mahajan et al, 2014). For example, genes that stimulate allelochemicals, enhance nutrient uptake and rapid growth habits can be transferred into crops (Haas and Streibig, 1982; Jabran, 2017).

In the Northern Guinea Savanna of Nigeria, under high weed pressure, NERICA1 and NERICA4 cultivarsproduce higher grain yield than CG14 (weed competitive but poor yielding) and ITA150, suggesting their tolerance to high weed pressure (Ekeleme et al, 2009). A similar trend is reported for NERICA1 in Southern Guinea Savanna (Ismaila et al, 2011; Kolo and Umaru, 2012). Regrettably, NERICA1 and other NERICA cultivarsare unable to prove similar outcomes in other neighbouring countries (Saito et al, 2012, 2018). Alternatively,IR74371-3-1-1 and Aus257 are reported as early maturing and high-yielding cultivars, with better weed-suppressive ability and greater nutrient uptake than NERICA1 (Saito, 2016). These cultivars are proposed to be better parents for breeding weed suppressive traits in upland rice (Saito and Futakuchi, 2014; Saito, 2016). It is therefore important to evaluate and update reports on potential cultivars that combine weed suppressive ability, higher yield and efficient nutrient uptake in low nutrient soils across upland rice ecologies in Nigeria.

Chemical weed control methods

The advantage of chemical weed control in upland rice compared to manual weeding in terms of higher yield and reduced labor cost has been reported over decades (Ogungbile and Lagoke, 1986; Akobundu, 1987; Lagoke et al, 1987; Usman, 2012; Adigun et al, 2017; Kolo et al, 2020). However, the performance of herbicide in the upland rice ecologies fluctuates, depending on the weed flora, weed growth stage and the pattern of weed emergence. Therefore, effective weed management in diverse field situations would require a combination of herbicide mixtures with a wide activity spectrum (Adigun et al, 2005). Table 2 shows potential herbicides or herbicide mixtures recommended for upland rice across agro-ecological zones in Nigeria.

Table 2. Herbicides recommended for different upland rice ecologies in Nigeria.

Pre-plant, Pre-plant emergence; PRE, Pre-emergence; POST, Post-emergence; NGS, Northern Guinea Savanna; FST, Forest transition; SW, South West; SGS, Southern Guinea Savanna.

In Guinea and Sudan Savanna zones of Nigeria, several pre-emergence (PRE) and post-emergence (POST) herbicides have been evaluated and recommended to farmers. Sole application of either glyphosate (pre-plant), oxadiargyl or butachlor has proven effective as PRE herbicides in the zones at recommended doses (Table 2). While POST herbicides are recommended either as preformulated mixture rates or mixed at specified doses to achieve better weed efficacy in the zones. For example, pretilachlor and pyribenzoxim mixtures, propanil and 2,4-D mixtures, and bispyribac sodium have been reported at specific recommended doses (Table 2).

In the rainforest zone of Nigeria (Table 2), only a few herbicide trials have been conducted compared with the Savanna ecological zone. Due to limited farm size and scarcity of tractors to plow, pre-plant/non- selective herbicides are often deployed to reduce the cost of plowing. Farmers often apply glyphosate and paraquat as pre-plant herbicides to clear weeds at the initial stage of cultivation. Although glyphosate and paraquat are not conventional rice herbicides, their broad-spectrum activity, acceptance and affordability have encouraged their consistent adoption as pre-plant herbicides under no-till upland and lowland rice production systems in the rainforest zones(Alagbo and Akinyemiju, 2018). For example, pre-plant application of glyphosate at 3.0 kg/hm2gives 70% weed control efficacy with an average grain yield of 3.9 t/hm2in a rainforest zone, and the cost of weed control with paraquat seems cheaper compared with hand weeding (Obadoni and Remison, 2007). This may partly explain the reasons why most farmers adopt paraquat above other herbicides in a mixed cropping system. Only a few PRE (i.e. butachlor) and POST (i.e. propanil + 2,4-D) herbicides have been evaluated in the zone so far (Table 2). New herbicides can be evaluated and recommended to improve weeding efficiency in the rainforest zone.

Table 3 shows herbicide classes recommended for weed control in upland rice over the past three decades in Nigeria. There is an urgent need to investigate the status quo of these rice herbicides in Nigeria. In sub-Saharan Africa, reasonable numbers of herbicides are absent among farmers and in the markets while unreported and/or unregistered new herbicide products presents in the market (Rodenburg et al, 2019). Increased demand for this product is largely attributed to high illiteracy levels among farmers in some sub-Saharan African countries (Achandi et al, 2018). For instance, in Nasarawa State of Nigeria, farmers rarely read herbicide labels, they depend largely on their neighbor’s knowledge of what herbicide is efficient to control weed in rice (Rodenburg et al, 2019). It therefore becomes necessary that favourite rice herbicides in the market are collected and evaluated in future research and appropriate dose recommended in different rice ecologies.

Herbicide-resistant rice

The adoption of herbicide-resistant rice has the potentials to reduce the proliferation of weeds that have developed resistance to conventional rice herbicides with time (Chauhan, 2013). For instance, imidazoline-resistant rice has been developed in the USA for this purpose (Gealy et al, 2003). Such technology may reduce overdependence on assorted rice herbicides in most upland ecosystems. Unfortunately, it is not deployed in Nigeria yet.

Integrated weed management strategies

Integrating two or more weed control methods is crucial for the sustainable management of weeds in upland rice culture in Nigeria (Tosh et al, 1981; Chauhan,2013; Adigun et al, 2017). Although herbicide application in upland rice has successfully reduced drudgeries/time wastage occasioned by manual weeding in the past three decades (Akobundu, 1987; Adeosun and Lagoke, 2005; Obadoni and Remison, 2007; Adigun et al, 2017), achieving season-long weed management is still difficult with herbicides, considering varying growth habits of weed species under changing climatic conditions (Adigun et al, 2005; Chauhan, 2012; Adigun et al, 2016). As temperature and atmospheric CO2levels begin to rise, below-ground parts, such as stolon and rhizomes, are predicted to spread fast, irrespective of water availability (Ramesh et al, 2017). Herbicides are predicted to become less effective (Mutti et al, 2019), in response to changes in anatomy, growth physiology and phenology of the target weed ?ora (Ramesh et al, 2017).Thus, farmers will have to carefully synchronize the timing of weed control with the weed life cycle (Ramesh et al, 2017). Addressing the aggressive weed pressure peculiar to upland rice ecosystems in Nigeria would necessitate the timely transfer of new and effective weed management technologies to complement existing weed management potentials in the farmers’ domain. Therefore, it is necessary that efficient/sustainable Integrated Weed Management (IWM) strategies should target possible reductions in the increasing spread of perennial propagules and heavy deposit of annual weed seeds in topsoil layers within upland ecosystems (Mahajan et al, 2014). Practical IWM strategies recommended for upland ecosystems in different agroecological zones are discussed under this section.

Table 3. List of herbicide classes recommended for upland rice in Nigeria in past three decades.

The different numbers indicate the groups each herbicide belongs according to the herbicide grouping of the Weed Science Society of America (WSSA, 2020).

2,4-D, 2,4-dichlorophenoxy acetic acid; ACCase, Acetyl CoA carboxylase inhibitor; EPSP, 5-enolpyruvylshikimate-3-phosphate.

In the Guinea Savanna ecosystem, the practice of conventional tillage is necessary to manage weeds effectively within a cropping cycle. This should be followed by manual or mechanical drilling/dibbing of rice seeds at 50–80 kg/hm2seeding rate, at 4–6 seeds per stand and 25 cm × 25 cm inter- and intra-row spacing (Table 4). For better enhancement of seedling growth, 300 kg/hm2of nitrogen, phosphorus and potassium (NPK) fertilizer formulation should be applied at 3 weeks after sowing (Enyinnia, 1992; Usman, 2012), such that it falls between the weed-free period to enhance nutrient uptake by rice seedlings rather than the weeds (Adeosun, 2005, 2008; Kolo et al, 2020). In a situation where soil nutrition is extremely low due to intensive cropping practices, about 10 t/hm2of poultry manure or other available sources of organic supplements could be applied as an alternative source of soil nutrient amendment (Danmaigoro et al, 2015). At a later growth stage (e.g. during panicle initiation), 150–200 kg/hm2urea should be applied to further enhance rice vegetative growth so that the canopy weed surpression effect can be sustained until harvest(Enyinnia, 1992; Usman, 2012).

Table 4. Integrated weed management prescriptions for rice upland ecological zones of Nigeria.

Either one of any of the listed post-emergence (POST) or pre-emergence (PRE) chemical methods can be integrated with other management strategies.

DAS, Days after sowing.

The following POST herbicides are expected to control weeds effectively when properly integrated with the above recommended strategies: propanil + 2,4-D formulation between 2.0–4.5 kg/hm2depending on weed pressure, pretilachlor (450 g/L) + pyribenzoxim (30 g/L) mixture and byspyribac sodium (40 g/L). Also, the following PRE herbicides have been recommended for upland rice: oxadiargyl (320 g/L), butachlor (1.5 kg/hm2) or a pre-plant glyphosate application where no-till is practised (Usman, 2012; Yawale et al, 2020).The practice of manual weeding operation is expected to supplement these herbicides either once or twice, especially where labor is cheap and affordable (Akobundu and Fagade, 1978; Adigun et al, 2017). Also, hand pulling should be adopted at a later growth stage in case of high weed infestation. When weed management strategies are well integrated, the following varieties show higher yield potentials in the Guinea Savanna zone: Ex-China (Dwarf), ITA257 (medium), FARO49 (IT315), FARO46 and NERICA1 (Adeosun et al, 1996; Ekeleme et al, 2009; Ismaila et al, 2011; Kolo and Umaru, 2012).

In the rainforest zone, conventional tillage is expected to precede any other operations, except for slopy topographic/hilly areas, where it becomes difficult to use tractors. The practice of direct seeding by drillings or dibbling at 5 seeds per hill at a spacing of 25 cm × 25 cm is equally established(Table 4). This is expected to be followed by the application of 200 kg/hm2NPK fertilizer at 42 d after sowing (DAS) (Enyinnia, 1992; Adigun et al, 2017). The only POST herbicide recommended for weed control in the rainforest zone is propanil + 2,4-D formulation (2.0 kg/hm2). The following PRE herbicides have also been recommended: butachlor (1.0 kg/hm2) (Enyinnia, 1992; Adigun et al, 2017), and a pre-plant application of glyphosate (3.0 kg/hm2) (Obadoni and Remison, 2007). Where noxious weed invasion persists, manual weeding is expected to supplement the efficacy of the above herbicides. For instance, the significant interaction between N and weed management under forest-savanna transition zone in Nigeria shows that yield response of upland rice to applied N is dependent on adopted/applied weed control method (Kolo et al, 2021). Application of N at 90 kg/hm2combined with the integration of butachlor and one manual weeding in a cropping cycle gives the rice a competitive advantage over weeds compared with either sole butachlor or manual weeding. Such integrated weed control efficiency could be attributed to the initial efficacy of the herbicide in inhibiting weed-root elongation and development (Agrawal et al, 2014), followed by subsequent weed removal by manual weeding before they mature (Kolo et al, 2021). However, limitationsof this technology could be the cost of labor and herbicides which are not considered in this review. ITA301, NERICA8 and FARO41 show significant yield potential in the rainforest zone under effective weed management (Enyinnia, 1992; Adigun et al, 2017).

PERSPECTIVES

Some forms of physical and cultural practices, such as tillage, manual weeding, use of weed competitive cultivars, enhanced N uptake by the rice crop and suitable cropping patterns, have been comprehensively studied with positive results across agro-ecological zones. However, their impact on crop yield under farmers’ fields seems insignificant. This may be due to farmers’ inadequate financial resources and scarcity of farm machinery when needed. The efficiency of cultural practices to suppress weeds in combination with other IWM approaches would depend largely on farmers’ knowledge of appropriate timing and availability of scarce resources. Hence, the availability of farm inputs like tractors, fertilizers, rice seeders, manual/mechanical weeders and herbicides should be prioritized. Also, farmers must be engaged with the technical know-how of IWM strategies.

It is no doubt that the effect of climate changes and prevalent agronomic practices (such as tillage operation and consistent herbicide application) have influenced a change in weed flora of upland ricefields over time. Therefore, a reassessment of weed flora composition, biology and ecology of problem weeds as affected by various agronomic practices in upland rice fields should be exhaustively reviewed. Most importantly, the interaction exists between competitive weed pressures and tillage practices in different rice ecologies.

The use of mechanical weeders to manage weeds looks promising in upland rice (Johnson et al, 2019). However, its efficiency, feasibility and acceptability among smallholder farmers should be further investigated in major rice agro-ecology.

Future research should show how different components of cultural practices could be harnessed and adopted for effective weed management in practical terms. For example, N fertilizer is better applied within the critical period of weed control to be efficiently utilized by rice other than the weeds (Emmanuel et al, 2021a; Kolo et al, 2021). However, it is uncertain whether such recommendations are strictly adhered to. Hence, it is need to identify their adoption levels and possible factors limiting its efficiency among smallholder farmers. Also, the use of weed competitive cultivars should be further extended to other rice agro-ecology. Introducing herbicide-resistant cultivars can help mitigate future resistance of weeds that have become tolerant to conventional rice herbicides in upland rice.

Several pre-plant (paraquat and glyphosate), PRE (butachlor and oxadiargl) and POST (propanil + 2,4-D amine, pretilachlor + pyribenzoxim and bispyribac sodium) herbicides have been tested and recommended for weed management in upland rice with different outcomes across agroecological zones. Nonetheless, herbicides and other weed management trials seem skewed to the savanna than the other ecological zones. To tackle the recent weed invasion, a multi-locational herbicide trial program is needed to select the best performing rice herbicides for different rice agro- ecologies. Also, the level of compliance with recommended herbicide technology among farmers remains uncertain. Many of the recommended rice herbicides in sub-Saharan Africa no longer exist in the market (Rodenburg et al, 2019). Therefore, it is necessary to validate rice herbicides of preference to rice farmers.

Achandi E L, Mujawamariya G, Agboh-Noameshie A R, Gebremariam S, Rahalivavololona N, Rodenburg J. 2018. Women’s access to agricultural technologies in rice production and processing hubs: A comparative analysis of Ethiopia, Madagascar and Tanzania., 60: 188–198.

Adeosun J O. 2005. Influence of nitrogen and period of weed interference on yield and yield components of upland rice () at Samaru Nigeria., 18: 45–55.

Adeosun J O. 2008. Predicting yield of upland rice (L.) under weed infestation., 21: 50–66.

Adeosun J O, Lagoke S T O. 2005. Evaluation of herbicide treatments for weed control in upland rice (L.) varieties., 18: 21–36.

Adeosun J O, Lagoke S T O, Chude V O, Ndahi W, Elemo K. 1996. Effectiveness of cinosulfuron for weed control in upland rice in the Northern Guinea Savanna of Nigeria., 9: 119–134.

Adeyemi O R, Atayese M O, Dare M O, Sakariyawo O S, Adigbo S O, Bakare T O. 2015. Weed control efficacy and arbuscular mycorrhizal (AM) colonization of upland rice varieties as affected by population densities., 5(1): 178–185.

Adigun J A, Lagoke S T O, Adekpe I D. 2005. Efficacy of selected herbicides for weed control in rainfed-upland rice in the Nigerian Northern Guinea Savanna., 38(3/4): 99–106.

Adigun J A, Osipitan A O, Lagoke S T, Adeyemi R O, Afolami S O. 2014. Growth and yield performance of cowpea ((L.) Walp) as influenced by row-spacing and period of weed interference in South-West Nigeria., 6(4): 188–198.

Adigun J A, Adeyemi O R, Lagoke S T O, Olorunmaiye P M, Daramola O S, Babatunde A O. 2016. Influence of inter-row spacing and weed control methods in groundnut (L.)., 16(1): 86–95.

Adigun J A, Kolo E, Adeyemi O R, Daramola O S, Badmus A A, Osipitan O A. 2017. Growth and yield response of upland rice to nitrogen levels and weed control methods., 11(6): 92–101.

Agrawal C, SenS, Singh S, Rai S, Singh P K, Singh V K, Rai L C. 2014. Comparative proteomics reveals association of early accumulated proteins in conferring butachlor tolerance in three N2-fixingspp., 96: 271–290.

Ajah J. 2014. Factors limiting small-scale farmers’ access and use of tractors for agricultural mechanization in Abuja, north central zone, Nigeria., 3(1): 115–124.

Akobundu I O. 1981. Weed control in direct-seeded lowland rice under poor water control conditions., 21(6): 273–278.

Akobundu I O. 1987. WeedScience in the Tropics: Principles and practices.New York, USA:John Wiley and Sons: 522.

Akobundu I O, Fagade S O. 1978. Weed problems in Africa rice lands.: Buddenhagen G I, Barsely T W. Rice in Africa. New York, USA: Academic Press: 10–20.

Alagbo O O. 2016. Effective weed control regime for lowland rice cultivation in southwestern Nigeria [Master Degree]. Ile-Ife, Nigeria: Obafemi Awolowo University.

Alagbo O O, Akinyemiju O A. 2018. Weed management practices by lowland rice farmers in the Southwest of Nigeria., 6(3): 1000372.

Ali A M, Sankaran S. 1984. Crop-weed competition in direct seeded flooded and rainfed bunded rice., 9(2): 22.

Biyi D. 2005. Government policies and competitiveness of Nigerian rice economy.: Rice Policy and Food Security in sub-Saharan Africa. Proceedings of a workshop held on 7–9 November 2005. Cotonou, Benin: 18.

Chauhan B S. 2012. Weed ecology and weed management strategies for dry-seeded rice in Asia., 26(1): 1–13.

Chauhan BS. 2013. Strategies to manage weedy rice in Asia., 48: 51–56.

Chauhan B S, Johnson D E. 2008a. Germination ecology of goosegrass (): An important grass weed of rainfed rice., 56(5): 699–706.

Chauhan B S, Johnson D E. 2008b. Germination ecology of southern crabgrass () and India crabgrass (): Two important weeds of rice in tropics., 56(5): 722–728.

Chauhan B S, Johnson D E.2008c. Germination ecology of two troublesome Asteraceae species of rainfed rice: Siam weed () and coat buttons ()., 56(4): 567–573.

Chauhan B S, Johnson D E. 2010. Effects of nitrogen on the competitiveness ofandwith direct seeded rice., 4: 14–19.

Dada O A, OladiranE M, Olubode O S, Togun A O. 2017a. Influence of weeding regimes on composition and diversity of weed species in upland rice (L.) field., 16(1): 62–74.

Dada O A, Oladiran E M, Togun A O. 2017b. Optimizing grain yield in upland rice field by synchronizing time of cattle dung application with period of weed removal in Southwestern Nigeria., 94(2): 141–156.

Danmaigoro O, Ishaya DB, Yakubu H. 2015. Performance of upland rice (L.) as influenced by weed control treatments, poultry manure and stand density., 12(2): 764–771.

Daramola O S, Adigun J A, Olorunmaiye P M. 2020. Challenges of weed management in rice for food security in Africa: A review., 53(3): 107–115.

Ekeleme F, Kamara A Y, Oikeh S O, Chikoye D, Omoigui L O. 2007. Effect of weed competition on upland rice production in north-eastern Nigeria.: Ahmed K Z. 8th African Crop Science Society Conference. 27–31 October. EI-Mina, Egypt: 61–65.

Ekeleme F, Kamara A Y, Oikeh S O, Omoigui L O, Amaza P, Abdoulaye T, Chikoye D. 2009. Response of upland rice cultivars to weed competition in the savannas of West Africa., 28(1): 90–96.

Emmanuel K, Aremu A J, Raphael A O, Samuel D O, Jacob B G. 2021a. The effect of weed control timing on the growth and yield of upland rice (L.)., 66(1): 27–38.

Emmanuel K, Aremu A J, Raphael A O, Samuel D O, Mojibade O P. 2021b. Growth and yield response of upland rice (L.) to different nitrogen fertilization and weeding levels., 86(2): 117–123.

Enyinnia T. 1992. Chemical weed control in rainfed upland rice in Nigeria., 38(4): 408–410.

FAO. 2014. A regional strategy for sustainable agricultural mechanization: Sustainable mechanization across agri-food chains in Asia and the pacific region.[2021-05-27] http://www. fao.org/.publications/card/en/c/78c1b49f-b5c2-43b5-abdf-e63bb6955f4f/.

Fischer A J, Ramírez H V, Gibson K D, da Silveira Pinheiro B. 2001. Competitiveness of semidwarf upland rice cultivars against palisadegrass () and signalgrass ()., 93(5): 967–973.

Franks S J, Sim S, Weis A E. 2007. Rapid evolution of flowering time by an annual plant in response to a climate fluctuation., 104(4): 1278–1282.

Gealy D R, Mitten D H, Rutger J N. 2003. Gene flow between red rice () and herbicide-resistant rice (): Implications for weed management., 17(3): 627–645.

Gill L S, Ene J C. 1978. Weed flora of the rice fields of Bendel State., 43: 182–183.

Haas H, Streibig J C. 1982. Changing patterns of weed distribution as a result of herbicide use and other agronomic factors.: LeBaron J, Gressel H M. Herbicide Resistance in PlantWiley New Jersey, USA: 57–80.

IRRI. 1973. Effect of Water Depth and Management on Grain Yield and Growth Characteristics of Broadcast Flooded Rice. Manila, the Philippines: International Rice Research Institute: 5–7.

Ismaila U, Kolo M G, Gbanguba A U. 2011. Effect of varieties and weeding regimes on weed and performance of two aerobic rice varieties., 1(3): 121–130.

Jabran K. 2017. Rice allelopathy for weed control.: Jabran K.Manipulation of Allelopathic Crops for Weed Control. Cham: Springer International Publishing: 35–47.

Jannink J L, Orf J H, Jordan N R, Shaw R G. 2000. Index selection for weed suppressive ability in soybean., 40(4): 1087–1094.

Johnson J M, Rodenburg J, Tanaka A, Senthilkumar K, Ahouanton K, Dieng I, Klotoe A, Akakpo C, Segda Z, Yameogo L P, Gbakatchetche H, AcheampongG K, Bam R K, Bakare O S, Kalisa A, Gasore E R, Ani S, Ablede K, Saito K. 2019. Farmers’ perceptions on mechanical weeders for rice production in sub-Saharan Africa., 55(1): 117–131.

Kamai N, Omoigui L O, Kamara A Y. 2020. Guide to Rice Production in Northern Nigeria. Ibadan, Nigeria: International Institute of Tropical Agriculture. http://www.iita.org.

Kamara A Y, Ekeleme F, Omoigui L O, Oikeh S O, Chikoye D, Tegbaru A. 2010. Response of upland rice cultivars to nitrogen fertilizer in the savannas of Nigeria., 102(1): 333–339.

Kehinde J K. 2002. Influence of seed rate and weed management practices on weed control and performance of upland rice., 15: 1–6.

Kolo E, Adigun J A, Adeyemi O R, Daramola O S, Bodunde J G. 2020. Effect of nitrogen levels and weed management methods on weed abundance and yield of upland rice (L.)., 65(2): 137–150.

Kolo E, Adigun J A, Daramola O S, Adeyemi O R, Olorunmaiye P M. 2021. Nitrogen effect on weed management in rice in a tropical forest-savanna transition zone of Nigeria., 35(2): 169–183.

Kolo M G M, Umaru I. 2012. Weed competitiveness and yield of inter- and intra-specific upland rice (L.) under different weed control practices at Badeggi, Niger State, Nigeria., 7(11): 1687–1693.

Kraehmer H, Jabran K, Mennan H, Chauhan B S. 2016. Global distribution of rice weeds: A review., 80: 73–86.

Lagoke S T O, Adejonwo K O, Falalu M M. 1987. Evaluation of directed post emergence or post-planting herbicide treatments for weed control in okra and chilli pepper mixed crops., 110: 94–95.

Mahadi M A, Dadari S A, Mahmud M, Babaji B A. 2006. Effect of pre-emergence herbicides on yield and yield components of rice., 4(2): 164–167.

Mahajan G, Chauhan B S, Kumar V. 2014. Integrated weed management in rice.: Chauhan G, Mahajan B S. Recent Advances in Weed Management. New York, USA: Springer: 125–153.

Miah M N H, Talukder S, Sarker M A R, Ansari T H. 2004. Effect of number of seedling per hill and urea supergranules on growth and yield of the rice cv. BINA Dhan4., 4(2): 122–129.

Mutti N K, Mahajan G, Jha P, Chauhan B S. 2019. The response of glyphosate-resistant and glyphosate-susceptible biotypes of junglerice () to mungbean interference., 67(4): 419–425.

Niang A, Becker M, Ewert F, Dieng I, Gaiser T, Tanaka A, Senthilkumar K, Rodenburg J, Johnson J M, Akakpo C, Segda Z, Gbakatchetche H, Jaiteh F, Bam R K, Dogbe W, Keita S, Kamissoko N, Mossi I M, Bakare O S, Cisse M, Baggie I, Ablede K A, Saito K. 2017. Variability and determinants of yields in rice production systems of West Africa., 207: 1–12.

Nojima K. 1996. Theory and practice of rice culture inter tillage.: Matsunayashi N, Ito M, Tarase R, Nomoto T, Yamada T. Theory and Practice of Growing Rice. Tokyo, Japan: Fuji Publishing: 396–399.

Obadoni B O, Remison S U. 2004. The weed flora of upland rice farms in Edo State, Nigeria., 17: 1–8.

Obadoni B O, Remison S U. 2007. Assessing the efficacy of some selected herbicides for weed control in upland rice production in Edo State, Nigeria., 6(2): 111–115.

Ogungbile A O, Lagoke S T O. 1986. On-farm evaluation of the economics of chemical weed control in oxen-mechanized maize production in Nigerian savanna., 32(4): 269–273.

Ogwuike P, Rodenburg J, Diagne A, Agboh-Noameshie A R, Amovin-Assagba E. 2014. Weed management in upland rice in sub-Saharan Africa: Impact on labor and crop productivity., 6(3): 327–337.

Olofintoye J A, Ajayi A S. 1997. Effects of land preparation practices and spacing on weed growth and yield of upland rice (L.)., 3: 31–37.

Phillip D, Jayeoba O O, Ndirpaya Y, Oluwole F. 2018. Innovation opportunities in the rice value chain in Nigeria., 2(3): 1–48.

Ramesh K, Matloob A, Aslam F, FlorentineS K, Chauhan B S. 2017. Weeds in a changing climate: Vulnerabilities, consequences, and implications for future weed management., 8: 95.

Rodenburg J, Johnson D E. 2009. Weed management in rice-based cropping systems in Africa.,103:149–218.

Rodenburg J, Saito K, Irakiza R, Makokha D W, Onyuka E A, Senthilkumar K. 2015. Labor-saving weed technologies for lowland rice farmers in sub-Saharan Africa., 29(4): 751–757.

Rodenburg J, Johnson J M, Dieng I, Senthilkumar K, Vandamme E, AkakpoC, Allarangaye M D, Baggie I, Bakare S O, Bam R K, Bassoro I, Abera B B, Cisse M, Dogbe W, Gbakatchétché H, Jaiteh F, Kajiru G J, Kalisa A, Kamissoko N, Sékou K, Kokou A, Mapiemfu-Lamare D, Lunze FM, Mghase J, Mossi Ma?ga I, Nanfumba D, Niang A, Rabeson R, Segda Z, Silas Sillo F, Tanaka A, Saito K. 2019. Status quo of chemical weed control in rice in sub-Saharan Africa., 11(1): 69–92.

Saito K. 2016. Plant characteristics of high-yielding upland rice cultivars in West Africa.,56(1): 276–286.

Saito K, Futakuchi K. 2014. Improving estimation of weed suppressive ability of upland rice varieties using substitute weeds., 162: 1–5.

Saito K, Azoma K, Rodenburg J. 2010. Plant characteristics associated with weed competitiveness of rice under upland and lowland conditions in West Africa., 116(3): 308–317.

Saito K, Sokei Y, Wopereis M C S. 2012. Enhancing rice productivity in West Africa through genetic improvement., 52(2): 484–493.

Saito K, Asai H, Zhao D L, Laborte A G, Grenier C. 2018. Progress in varietal improvement for increasing upland rice productivity in the tropics.,21(3): 145–158.

Saito K, Vandamme E, Johnson J M, Tanaka A, Senthilkumar K, Dieng I, Akakpo C, Gbaguidi F, Segda Z, Bassoro I, Lamare D, Gbakatchetche H, Abera B B, Jaiteh F, Bam R K, Dogbe W, Sékou K, Rabeson R, Wopereis M C S. 2019. Yield-limiting macronutrients for rice in sub-Saharan Africa., 338: 546–554.

Sankaran S, deDatta S K. 1986. Weeds and weed management in upland rice., 38: 283–337.

Singh B N, Fagade S O, Ukungwu M N, William C, Jagtap S S, Oladimeji O, Effisue A, Okhidievie O. 1997. Rice growing environments and biophysical constraints in different agroecological zones of Nigerian., 2: 35–41.

Singh R S, Ghosh D C. 1992. Effect of cultural practices on weed management in rainfed upland rice., 38(2): 119–121.

Singh S, Ladha J K, Gupta R K, Bhushan L, Rao A N, Sivaprasad B, Singh P P. 2007. Evaluation of mulching, intercropping with Sesbania and herbicide use for weed management in dry-seeded rice (L.)., 26(4): 518–524.

Tosh G C, Reddy B S, Nanda K C. 1981. Integrated weed management on dryland rice., 6(5): 20.

Usman A, Elemo K A, Misari S M, Lagoke S T O, Adigun J A. 2002. Nitrogen and weed management in maize intercropped with upland rice., 21(1): 5–16.

Usman I. 2012. Evaluation of post emergence herbicides on weed control performance and profitability of rice () at Lafiagi, Kwara State of Nigeria., 3(5): 236–240.

WSSA (Weed Science Society of America). 2010. Common and chemical names of herbicides approved by the Weed Science Society of America., 58(6): 511–518.

Yawale M A, Hussaini M A, Garko M S, Lado A, Muhammad A. 2019. Predicting critical period of weed control for optimum yield of upland rice (L.) in the Sudan savanna of Nigeria., 12(9): 69–74.

Yawale M A, Garko M S, Sa’ad A M, Bello T T. 2020. Effect of weed phytosociological characteristics on control practices in upland rice production under Sudan savanna of Nigeria., 7(2): 273–280.

Zhao D L, Atlin G N, Bastiaans L, Spiertz J H J. 2006. Cultivar weed-competitiveness in aerobic rice: Heritability, correlated traits, and the potential for indirect selection in weed-free environments., 46(1): 372–380.

Ziska L H. 2003. Evaluation of the growth response of six invasive species to past, present and future atmospheric carbon dioxide., 54: 395–404.

5 September 2021;

9 November 2021

Oyebanji O. Alagbo (oyebanjialagbo@oauife.edu.ng)

Copyright ? 2022, China National Rice Research Institute. Hosting by Elsevier B V

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Peer review under responsibility of China National Rice Research Institute

http://dx.doi.org/10.1016/j.rsci.2021.11.004

(Managing Editor: Wang Caihong)