Effects of grain-producing cover crops on rice grain yield in Cabo Delgado, Mozambique

Plantas de cobertura produtoras de grãos e a produtividade de arroz em Cabo Delgado, Moçambique

ABSTRACT

Besides providing benefits to the environment such as soil protection, release of nutrients, soil moisture maintenance, and weed control, cover crops can increase food production for grain production. The aim of this study was to evaluate the production of biomass and grain cover crops (and its respective effects on soil chemical and physical attributes), yield components, and grain yield of rice in Mozambique. The study was conducted in two sites located in the province of Cabo Delgado, in Mozambique. The experimental design was a randomized block in a 2 × 6 factorial, with four repetitions. Treatments were carried out in two locations (Cuaia and Nambaua) with six cover crops: Millet (Pennisetum glaucum L.); namarra bean (Lablab purpureus (L.) Sweet), velvet beans (Mucuna pruriens L.), oloco beans (Vigna radiata (L.) R. Wilczek), cowpea (Vigna unguiculata L.), and fallow. Cover crops provided similar changes in chemical and physical properties of the soil. Lablab purpureus, Vigna unguiculata, and Mucuna pruriens produced the highest dry matter biomass. Vigna unguiculada produced the highest amount of grains. Rice grain yields were similar under all cover crops and higher in Cuaia than Nambaua.

Key words:
sustainability; legumes; grain production; conservation agriculture

RESUMO

As plantas de cobertura, além de proporcionar benefícios ao ambiente como a proteção do solo, liberação de nutrientes, manutenção da umidade do solo e controle de plantas daninhas, pode servir para aumentar a produção de alimentos pela sua própria produção de grãos. O objetivo desse trabalho foi avaliar a produção de biomassa seca e de grãos de espécies de plantas de cobertura e seu respectivo efeito nos atributos químicos e físicos do solo, componentes de produção e produtividade de arroz em Moçambique. O trabalho foi desenvolvido em dois locais na Província de Cabo Delgado, em Moçambique. O delineamento experimental foi em blocos ao acaso no esquema fatorial 2 x 6, com 4 repetições. Os tratamentos foram compostos pelas 2 localidades (Cuaia e Nambaua) e 6 coberturas vegetais: Milheto (Pennisetum glaucum L.); Lablabe (Lablab purpureus (L.) Sweet), mucuna (Mucuna pruriens L.), feijão oloco (Vigna radiata (L) R. Wilczek), feijão caupi (Vigna unguiculata L.) e pousio. As plantas de cobertura proporcionaram alterações semelhantes nas propriedades químicas e físicas do solo. Lablab purpureus, Vigna unguiculata e Mucuna pruriens produziram a maior quantidade de biomassa seca. Vigna unguiculata foi a cultura de cobertura que produziu a maior quantidade de grãos. A produtividade de grãos do arroz foi semelhante em todas as culturas de cobertura e maior em Cuaia que Nambaua.

Palavras-chave:
sustentabilidade; legumes; produção de grãos; agricultura de conservação

INTRODUCTION

About 270 million people living in Sub-Saharian Africa face problems of hunger and malnutrition (Balasubramanian et al., 2007). Despite having climate and soil conditions to produce their own food, countries of this region are characterized by being importers of foodstuff (Kijima et al., 2011; Benson et al., 2008; Ivanic & Martin, 2008). Therefore, there is a need for developing actions and technologies that effectively contribute to the increase of food production in this area.

Rice (Oryza sativa L.) is considered a staple food for countries worldwide (Nokkoul & Wichitparp, 2014; Nascente et al., 2013a; Nayar, 2014). Specifically in Mozambique, this grain can ease poverty of 3.1 million people directly dependent on rice grain production and 20 million Mozambicans indirectly dependent on it (IRRI, 2015). In 2010, rice was produced on 185,000 ha, resulting in a production of 180,000 tons of paddy rice in this country. However, this production was not enough to meet people’s demand, which led to the importation of about 304,000 metric tons of rice (Ricepedia, 2015). Therefore, it seems that, despite the fact that climate and soil conditions are favorable for rice production, the yield is very low, ranging from 970 kg ha^-1 (Ricepedia, 2015) to 1170 kg ha^-1 (Faostat, 2015). The main reasons are the use of rudimentary techniques, limited knowledge, and inefficient management of water and infrastructure, which keeps rice production in Mozambique, and in several African countries, in family subsistence levels (Balasubramanian et al., 2007; Planeta Arroz, 2015). Thus, there is a need for developing farming techniques in these sites to take advantage of climate and soil favorable conditions and provide significant increases in the yield of rice grains.

The no-tillage system (NTS) can be a viable alternative to Mozambique and other countries in Africa, Latin America, and Asia. It is a technique that allows several environmental benefits, such as increasing levels of organic matter and biological activity of the soil, reduction of soil temperature fluctuations and laminar erosions. In addition, it reduces the carrying of fertilizers and pesticides to the watershed water, reductes the population of weeds and allows greater conservation of soil moisture, being therefore considered a sustainable production technique (Nascente et al., 2013b).

One of the premises of the NTS is the use of cover crops to form straw layers on the soil surface prior to deployment of the main crop (Crusciol et al., 2015). Among many benefits, cover cops can significantly affect the chemical properties of the soil (Boer et al, 2007; Carpim et al., 2008; Garcia et al., 2008; Torres & Pereira, 2008; Cunha et al., 2011; Nascente et al., 2015). Sá (1993) reported the use of cover crops for 16 years and the increase in phosphorus concentration from 29 to 129 mg kg^-1. Sá (1993), Crusciol et al. (2015), and Nascente et al. (2015) showed increases in K⁺ levels in NTS because of the use of cover crops. Other studies also showed accumulations of Ca²⁺, Mg²⁺ (Falleiros et al., 2003), Zn²⁺, Mn²⁺, Fe²⁺, and Cu²⁺ (Franzluebbers & Hons, 1996), increases in cation exchange capacities, in soil organic matters, and in P and K⁺ (Crusciol et al., 2015; Nascente et al., 2015), and also changes in pH and reductions of Al saturation (Cunha et al., 2011) due to cover crops in NTS.

The inclusion of cover crops before rice cultivation can contribute to sustainable development (Filizadeh et al., 2007; Nascente et al., 2013a). The increased diversity of plant species in the environment provides benefits such as better use of soil nutrients, minor attacks of pests, lower incidence of pathogens, greater weed control, increased crop yields, and greater yield stability (Mahmoudi et al., 2011; Yahuza, 2011).

The use of cover crops can also serve as a source of food for farmers, especially when using grain-producing species (Filizadeh et al., 2007). This feature is very important for family farmers - especially in Latin America, Africa, and Asia - that face problems of malnutrition and starvation, once they can enrich the family diet by including more food without the need to enlarge the area. However, considering the peculiarities of each region, there is still little knowledge on the main characteristics of cover crops with the production of biomass and grains (Oliveira et al., 2002). In this sense, the objective of this study was to evaluate the production of biomass and grains by cover crops and its effects on soil chemical and physical properties, yield components, and grain yield of rice in Mozambique.

MATERIAL AND METHODS

The study was conducted in two districts of the Province of Cabo Delgado, in Mozambique, namely Metuge and Macomia. In Metuge, the trial was carried out in an experimental field in the village of Cuaia (40°23'47" E and 12°58'17" S), while in the district of Macomia, it was carried out in the village of Nambaua (40°25'54" E and 11°45'11" S). The climate is Tropical Savanna, considered as Aw according to Köppen’s classification. There are two distinct seasons, usually the dry season from May to September (fall/winter) and the rainy season from October to April (spring/summer). Cuaia has an altitude of 100 m and average annual rainfall varies from 800 to 1000 mm, with an average annual temperature of 25 °C. In Nambaua, the altitude is 1000 m, the average annual rainfall varies from 1000 to 1300 mm, and the average temperature is 22 °C. Besides, from the “Instituto Nacional de Meteorologia de Moçambique (INAM)” we got information from medium temperature and rain in both sites (Figures 1 and 2). The soil in Cuaia is classified as Ferralic arenosol (FAO, 2014). The values of the soil texture in the 0-0.20 m layer were 672.7 g kg^-1 sand, 148.3 g kg^-1 silt, and 179.0 g kg^-1 clay. In Nambaua, the soil is classified as Haplic lixisol (FAO, 2014). The values of the soil texture in the 0-0.20 m layer were 728.3 g kg^-1 sand, 106.1 g kg^-1 silt, and 165.6 g kg^-1 clay.

Figure 1:
Mean temperature and rain during the trial period in the Cuaia Site.

Figure 2:
Mean temperature and rain during the trial period in the Nambaua Site.

Before the application of treatments in March 2014, the chemical characteristics of the soil were determined according to the methods described by Donagema et al. (2011). The results for Cuaia were: pH (H₂O) = 6.35, Ca²⁺ = 5.8 cmol_c kg^-1, Mg²⁺ = 4.2 cmol_c kg^-1, K⁺ = 0.70 cmol_c kg^-1, H⁺ + Al³⁺ = 0.64 cmol_c kg^-1, Na⁺ = 0.69 cmol_c kg^-1, cmol_c kg^-1, P = 9.25 mg kg^-1, and soil organic matter = 27.7 g kg^-1. The results for Nambaua were: pH (H₂O) = 6.21, Ca²⁺ = 5.3 cmol_c kg^-1, Mg²⁺ = 6.5 cmol_c kg^-1, K⁺ = 1.20 cmol_c kg^-1, H⁺ + Al³⁺ = 0.87 cmol_c kg^-1, Na⁺ = 0.30 cmol_c kg^-1, P = 40.83 mg kg^-1, and soil organic matter = 17.0 g kg^-1.

The experimental design was a randomized block in a 2 × 6 factorial with four repetitions. Treatments were composed of two environments (Cuaia and Nambaua) and six vegetation covers: Pennisetum glaucum L., Lablab purpureus (L.) Sweet, Mucuna pruriens L., Vigna radiata (L.) R. Wilczek, Vigna unguiculata L., and fallow. The plots had the dimension of 2 × 5 m. The usable area was considered the two central rows of the plot, disregarding 0.50 m by the end on each side of both plots following methods proposed by Nascente et al. (2013b). There was a 1 m wide alley between each plot.

Cover crops were sown on April 30^th, 2014 in Cuaia, and on July 18^th, 2014 in Nambaua (this area was flooded and then we had to wait the natural drainage before sowing cover crops). The row spacing used was of 0.40 m between plants with density of 10 seeds per meter. Fertilization was not carried out in the plots. The weed control was manual with the aid of hoes; there were weed problems (Cyperus rotundus L.) especially in Nambaua,. Control of insect plagues and diseases was not performed. Harvest of cover crops took place on August 30^th in Cuaia and on December 5^th in Nambaua. The grains were manually harvested and the aboveground plants were cut and placed to dry until reaching constant weight. Data concerning the production of dry biomass and grains of cover crops were collected. After weighing the dry matter, biomass was placed on the soil surface in the respective plots.

In both locations, rice cultivar Macassane was sown in nursery (January 1^st, 2015 in Cuaia and January 2^nd, 2015 in Nambaua) and manually transplanted 15 days after emergence, with a space of 0.40 m between rows and 0.20 m between plants. Cultivar Macassane was developed by the Mozambique Agricultural Research Institute (IIAM); this cultivar, very known by the Mozambique farmers, has a life cycle of around 90 days, low plant height (< 0.90 m), and long aromatic grain (IIAM, 2015). No fertilizations and control of insect plague and diseases were implemented. Weed control was carried out by weeding; in Nambaua, there was a great infestation of Cyperus rotundus.

Rice harvesting in both locations was done manually after physiological maturity (April 4^th, 2015 in Cuaia and April 6^th, 2015 in Nambaua). The following data were collected in each plot: number of tillers, determined by measuring 10 plants per plots at the full flowering stage; plant height (m) was determined by measuring 10 plants per plot at the time when the crop was at the phenological stage of pasty grains and, by recording the distance between the soil surface and the top end of the highest panicle; number of panicles m^-1 was determined by counting the number of panicles within 1.0 linear meter of one of the rows in the useful area of each plot; number of grains panicles^-1 was determined by counting the number of grains in 10 randomly collected panicles and by calculating their average; mass of 1000 grains was randomly evaluated by collecting and weighing two samples of 1000 grains from each plot and then corrected to 13% of water content; and grain yield was determined by weighing the harvested grains in the usable area of each plot, corrected to 13% of water content, and converted to kg ha^-1.

The soil was sampled before sowing cover crops on April 3^rd, 2014 in Nambaua and on April 2^nd, 2014 in Cuaia and before rice transplanting on December 22^nd, 2014 in Nambaua and December 20^th, 2014 in Cuaia. Eight single soil samples were collected from each plot in the 0-0.20 m layer, being manually mixed and homogenized to form a composite sample of each plot. These samples were packed separately in plastic bags and sent for chemical and physical analysis, according to the methodology of Embrapa (Donagema et al., 2011).

The pH was determined in water by using a soil solution ratio of 1:2.5. Phosphorus and K⁺ were extracted by Mehlich-1, Ca²⁺, Mg²⁺, Na+, and Al³⁺ with 1 mol L^-1 KCl. In the extracted solution, P was determined by colorimetry and K⁺ by flame photometry. Ca²⁺ and Mg²⁺ were determined by EDTA titration and Al³⁺ by NaOH titration from the extract. Soil organic matter was determined by Walkley & Black’s method (Walkley & Black, 1934).

Soil bulk density of the soil was calculated by the relation between the mass and the volume of the soil using the volumetric ring (50 cm³). Soil texture (sand, silt, and clay) was calculated by the Pipette method.

Variance analysis was performed with the data of physical and chemical soil properties, dry biomass and grain production of cover crops, plant height, yield components, and grain yield of rice. Cover crops and locations were considered fixed effects, while blocks and interactions were considered random effects. In the data with significant effects, a mean comparative Tukey test was carried out at P < 0.05, using the statistical package SAS (SAS, 1999).

RESULTS

In Cuaia, no differences were found among cover crops for chemical characteristics values of pH, Na, Ca, Mg, H + Al, K, P, and soil organic matter (Table 1), and neither for the physical attributes of the soil: particle density, sand, silt, and clay contents. In Nambaua, no differences were found among cover crops for chemical and physical properties of the soil (Table 2).

Table 1:
Soil chemical and physical attributes at cover crops sowing day (CCD) and at rice transplantation day (RTD) at the 0-0.20 m layer as a function of cover crops (CC). Cuaia site, Province of Pemba, Moçambique

Chemical attribute
CC1	pH (water)		Na (cmolc kg-1)		Ca (cmolc kg-1)		Mg (cmolc kg-1)
	CCD	RTD	CCD	RTD	CCD	RTD	CCD	RTD
	Average		Average		Average		Average
1	6.24		0.69		6.88		5.39
2	6.51		0.70		6.80		3.75
3	6.35		0.69		6.57		3.67
4	6.52		0.91		5.71		4.14
5	6.28		0.88		5.16		4.38
6	6.18		0.80		6.80		2.97
Average	6.44	6.25	0.76	0.80	5.99	6.64	4.69	3.41
	H+Al (cmolc kg-1)		SOM (g kg-1)		K (mg kg-1)		P (mg kg-1)
1	0.71		25.50		0.97		19.38
2	0.70		23.49		1.05		20.00
3	0.72		25.56		0.96		19.38
4	0.73		25.48		0.84		10.88
5	0.84		28.05		1.09		9.75
6	0.65		25.69		0.94		15.00
Average	0.73	0.72	25.97	25.28	0.96	0.99	14.95	16.50
Factor	ANOVA (F probability)
	pH	Na	Ca	Mg	H+Al	MO	K	P
CC	0.5388	0.5018	0.4081	0.8422	0.7154	0.8575	0.6315	0.7249
Time (T)	0.1508	0.0589	0.2613	0.2303	0.0784	0.7217	0.0574	0.0963
CC × T	0.1576	0.5883	0.9528	0.6765	0.4173	0.1905	0.1829	0.7312
Physilcal attribute
CC	SBD (g kg-1)		Sand (g kg-1)		Silt (g kg-1)		Clay (g kg-1)
	CCD	RTD	CCD	RTD	CCD	RTD	CCD	RTD
	Average		Average		Average		Average
1	1.26		712.5		145.0		142.5
2	1.24		725.0		157.5		118.8
3	1.31		728.8		145.0		123.8
4	1.36		741.2		97.50		160.0
5	1.29		755.0		113.8		132.5
6	1.24		662.5		167.5		172.5
Average	1.27	1.28	682.5	759.2	166.7	108.8	150.8	132.5
Factor	ANOVA (F probability)
	SBD		Sand		Silt		Clay
CC	0.5961		0.7164		0.6220		0.2858
Time (T)	0.8574		0.0629		0.0687		0.2319
CC × T	0.5074		0.3225		0.1915		0.0897

• ¹
  1 - millet (Pennisetum glaucum L.); 2 - namarra beans (Lablab purpureus (L.) Sweet), 3 - velvet bean (Mucuna pruriens L.); 4 - oloco beans (Vigna radiata (L.) R. Wilczek); 5 - cowpea (Vigna unguiculata L.); and 6 - fallow.
• SOM - soil organic matter; SBD - soil bulk density.

Table 2:
Soil chemical and physical attributes at cover crops sowing day (CCD) and at rice transplantation day (RTD) at the 0-0.20 m layer as a function of cover crops (CC). Nambaua site, Province of Pemba, Moçambique

Chemical attribute
CC1	pH (water)		Na (cmolc kg-1)		Ca (cmolc kg-1)		Mg (cmolc kg-1)
	CCD	RTD	CCD	RTD	CCD	RTD	CCD	RTD
	Average		Average		Average		Average
1	6.17		0.35		5.00		4.22
2	6.44		0.45		5.32		4.92
3	6.18		0.58		5.08		3.60
4	6.31		0.52		5.08		6.02
5	6.25		0.44		3.60		3.91
6	6.23		0.37		4.30		5.94
Average	6.16	6.37	0.42	0.48	4.92	4.44	5.11	4.43
	H+Al (cmolc kg-1)		SOM (g kg-1)		K (mg kg-1)		P (mg kg-1)
1	0.83		18.96		0.81		38.00
2	0.84		23.69		1.04		39.88
3	0.83		22.79		0.87		38.25
4	0.74		21.53		0.85		40.00
5	0.63		18.96		0.91		34.38
6	0.79		22.68		0.80		51.00
Average	0.79	0.76	21.32	21.58	0.90	0.86	43.21	37.29
Factor	ANOVA (F probability)
	pH	Ca	Mg	Na	H+Al	K	P	MO
CC	0.7558	0.0701	0.6298	0.4729	0.2934	0.8282	0.6118	0.7836
Time (T)	0.0722	0.0604	0.5088	0.0847	0.0682	0.1541	0.1102	0.1816
CPC × T	0.8010	0.1379	0.7829	0.7053	0.8698	0.7918	0.6066	0.5990
Physical attribute
CC1	SBD (g kg-1)		Sand (g kg-1)		Silt (g kg-1)		Clay (g kg-1)
	CCD	RTD	CCD	RTD	CCD	RTD	CCD	RTD
	Average		Average		Average		Average
1	1.22		755.0		107.5		136.3
2	1.17		728.8		145.0		125.0
3	1.17		750.0		117.5		135.0
4	1.20		737.5		110.0		152.5
5	1.21		776.3		112.5		111.3
6	1.19		780.0		93.8		122.5
Average	1.21 A	1.19 A	734.5 A	774.6 A	102.9 A	125.8 A	162.0 A	98.8 A
Factor	ANOVA (F probability)
	PD		Sand		Silt		Clay
CC	0.7878		0.6037		0.4351		0.2362
Time (T)	0.1059		0.0585		0.1086		0.0894
CC x T	0.7674		0.9807		0.6585		0.6154

• ¹
  1 - millet (Pennisetum glaucum L.); 2 - namarra beans (Lablab purpureus (L.) Sweet), 3 - velvet bean (Mucuna pruriens L.); 4 - oloco beans (Vigna radiata (L.) R. Wilczek); 5 - cowpea (Vigna unguiculata L.) and 6 - fallow.
• SBD - soil bulk density; SOM - soil organic matter.
• Means followed by the same letter, uppercase vertically or lowercase horizontally, do not differ by Tukey test at p ≥ 0.05.

There was a significant interaction between cover crops species and locations regarding dry biomass (Table 3). Concerning the production of grains, only effects of cover crops species were found. Vigna unguiculata achieved the highest grain yield (1793 kg ha^-1) and differed from all other species.

Table 3:
Biomass dry matter (BDM) and grain yield (YIELD) of cover crops cultivated before rice crops. Cuaia and Nambaua sites, Pemba city, Province of Cabo Delgado Moçambique, Growing season 2014

Factor	BDM	YIELD
Cover crop	kg ha-1
Pennisetum glaucum	1000 b	238 c
Lablab purpureus	3188 a	700 b
Mucuna pruriens	3213 a	800 b
Vigna radiata	350 b	209 c
Vigna unguiculata	3163 a	1793 a
Site
Cuaia	1815 b	780 a
Nambaua	2550 a	716 a
Factor	ANOVA (F probability)
Cover crop (CC)	< 0.001	0.0307
Site	0.0109	0.9027
CC× site	0.0271	0.5548

• *Means followed by the same letter vertically, do not differ by the Tukey test at p ≥ 0.05.

In the interaction between cover crops and locations in relation to dry biomass of cover crops, Pennisetum glaucum, Lablab purpureus, Mucuna pruriens, and V. radiata produced more in Nambaua than in Cuaia (Table 4). In Cuaia, V. radiata had the highest biomass production and differed from the other cover crops. In Nambaua, L. purpureus produced more biomass and differed from the other cover crops.

Table 4:
Interaction between biomass dry matter of cover crops and sites. Cuaia and Nambaua sites, Pemba city, Province of Cabo Delgado Moçambique, Growing season 2014

Factor	Cuaia	Nambaua
Cover crop	kg ha-1
Pennisetum glaucum	875 cdB	1125 dA
Lablab purpureus	2000 bcB	4375 aA
Mucuna pruriens	2750 bB	3675 bA
Vigna radiata	200 dB	700 dA
Vigna unguiculata	3450 aA	2875 cB

• *Means followed by the same letter, uppercase horizontally or lowercase vertically, do not differ by the Tukey test at p ≥ 0.05.

Cover crops did not significantly affect tillering, plant height, number of panicles and number of grains per panicle, mass of 1000 grains, and grain yield of rice (Table 5). On the other hand, there were effects of cropping location on tillering, plant height, number of panicles, number of grains per panicle, and grain yield and for all these variables and, values were higher in Cuaia than in Nambaua.

Table 5:
Plant height (PH), number of tillers (NT), number of panicles (PAN), number of grains per panicle (GRAIN), mass of 1000 grains (MGRAIN), and grain yield of rice as affected by cover crops and sites. Cuaia and Nambaua sites, Pemba city, Province of Cabo Delgado Moçambique, Growing season 2014/2015

Cover crop	PH	NT	PAN	GRAIN	MGRAIN	YIELD
	cm	n. m-1	n. m-1	n. panicle-1	grams	kg ha-1
Pennisetum glaucum	56.1	18.5	14.0	141	24.7	3248
Lablab purpureus	55.0	20.6	13.9	158	23.8	3499
Mucuna pruriens	55.4	20.6	13.4	151	25.5	3586
Vigna radiata	57.9	21.5	16.3	147	25.6	3700
Vigna unguiculata	62.6	18.6	13.8	146	25.3	3816
Fallow	60.4	20.4	13.3	145	23.9	3586
Site
Cuaia	74.7 a	22.0 a	15.3 a	158 a	24.9 a	4509 a
Nambaua	40.3 b	18.0 b	12.9 b	139 b	24.8 a	2594 b
Factor	ANOVA (F probability)
Cover crops (CC)	0.8758	0.8724	0.2055	0.7925	0.1368	0.8545
Site	< 0.0001	0.0231	0.0024	0.0127	0.9835	< 0.001
CC×Site	0.7905	0.2963	0.1669	0.1629	0.2411	0.3655

• *Means followed by the same letter vertically do not differ by the Tukey test at p ≥ 0.05.

DISCUSSION

Cover crop plants were grown for a period of 120 days in Cuaia and for 137 days in Nambaua. Based on the results, we can infer that the period for development of cover crops was not enough to provide significant differences in chemical and physical properties of the soil in relation to the control treatment (fallow). Nevertheless, the use of cover crops may be advantageous since the fallow treatment favors the spread of weeds (Nascente et al., 2013b).

Cover crops produced low amounts of dry biomass, between 350 and 3213 kg ha^-1 on average, in both locations (Table 3); therefore, we can assume that the few months of growing cover crops was not enough to provide significant differences in soil properties. Nascente et al. (2015) observed significant increases in soil nutrient levels, soil organic matter, cation exchange capacity, and base saturation after two years of cultivating cover crops by following rice crops. Cover crops such as grasses Panicum maximum, Brachiaria ruziziensis, and Brachiaria brizantha produce large amounts of dry biomass (> 10 Mg ha^-1) and provide significant increases in soil fertility (Moreti et al., 2007; Rosolem et al., 2010; Pacheco et al., 2011; Nascente et al., 2015). However, despite not having significant changes in the chemical and physical properties of the soil, due to the use of cover crops in our trial, this practice should be continued because, ,on the long run, it can provide cycling nutrients and other benefits such as maintenance of soil moisture, protection against erosion, and lower oscillation of soil temperature (Crusciol et al., 2015; Nascente et al., 2015) and can produce grains. Moreti et al. (2007), Cunha et al. (2011), and Nascente et al. (2015) added that by using cover crops, the chemical properties of soil are enhanced with their regular use, i.e., it is interesting to use cover crops every year before rice cultivation.

It is important to highlight that we did not use any fertilization. However, it is important to supply the soil with nutrients that were removed in the harvesting, especially with cash crops to avoid reductions in soil fertility. In this sense, soil fertilization with the use of cover crops would help to increase nutrient efficiency once nutrients lost by leaching could be uptaken by cover crop roots and returned to the topsoil after harvesting (Crusciol et al., 2015).

From our results, Vigna unguiculata stood out because, on average, it produced a higher amount of biomass in relation to the other cover crops and produced the highest amount of grains that can be used as food by family farmers. According to Andrade Junior et al. (2002), V. unguiculata has a short cycle, low water requirement, and rusticity to develop in low fertility soils and, its grain is an excellent source of protein, carbohydrates, vitamins, and minerals. Because of these features, V. unguiculata becomes an important alternative for food production preceding rice cultivation.

Plants differ in developing specific conditions (Oliveira et al. (2002), being, therefore, interesting to evaluate different locations. Based on the results, we found that V. unguiculata grew more in Cuaia. On the other hand, Lablab purpureus and Mucuna pruriens grew better in Nambaua. This could be because in Nambaua, the soil had 34.38 mg kg^-1 of P and in Cuaia, 9.75 mg kg^-1. Therefore, it is likely that the other cover crops develop better in places with high level of P. However, Vigna unguiculata, among the cover crops tested, seems to have higher response to K in the soil, once in Cuaia, the level of this nutrient was higher than in Nambaua. According to Oliveira et al. (2009), the cover crop V. unguiculata increased plant growth and grain yield with the increase of K fertilization.

Considering that productivity of rice grains is determined by three yield components, namely the number of panicles m^-2 and of filled grains and the mass of 1000 grains (Yoshida, 1981), from the results obtained in the yield components, we can explain the higher grain yield in Cuaia compared with Nambaua.

In both sites, rice grain yields were high considering the yield average in Mozambique in the growing season of 2013, 1170 kg ha^-1 (Faostat, 2015). Therefore, grain yield obtained in Nambaua was more than twice the national average and grain yield achieved in Cuaia was almost four times higher than the rice grain yield in Mozambique. These data are promising and may contribute incisively to increase food production in Africa and could be used in other countries of Asia and Latin America. Allied to this, there is the production of grains from the cover crops that can be grown before rice cultivation. Therefore, aiming to achieve a sustainable development by including more species in agricultural systems, we observed that the use of cover crops, especially V. unguiculata, followed by rice cultivation, could contribute to increase grain production (from cover crop and from rice), resulting in much higher food per area. Furthermore, cover crops provided nutrient cycling, which could allow better development of the following rice crop. These data become more important considering that around 270 million of inhabitants living in sub-Saharan Africa have problems with hunger and malnutrition (Sanchez & Swaminathan, 2005; Balasubramanian et al., 2007).

CONCLUSION

Cover crops and fallow resulted in similar chemical and physical properties of soil and rice grain yield;

Lablab purpureus, Vigna unguiculata, and Mucuna pruriens provide higher production of biomass;

Vigna unguiculada produce the highest grain yield;

The location of Cuaia allows higher rice grain yield than the location of Nambaua.

ACKNOWLEDGEMENTS

The authors would like to thank the MKTPlace plataform for financial support (ID: 532) and CNPq for the research productivity scholarship granted to the first author.

REFERENCES

• Andrade Júnior AS, Santos AA, Athayde Sobrinho C, Bastos EA, Melo FB, Viana FMP, Freire Filho FR, Carneiro JS, Rocha MM, Cardoso MJ, Silva PHS & Ribeiro VQ (2002) Cultivo do feijão-caupi [Vigna unguiculata (L.) Walp.]. Teresina, Embrapa Meio-Norte. 108 p.
• Balasubramanian V, Sie M, Hijmans RJ & Otsuka K (2007) Increasing rice production in sub-Saharan Africa: Challenges and opportunities. Advances in Agronomy, 94:55-133.
• Benson T, Mugarura S & Wanda K (2008) Impacts in Uganda of rising global food prices: The role of diversified staples and limited price transmission. Agricultural Economics, 39:513-524.
• Boer CA, Assis RL, Silva GP, Braz AJBP, Barros ALL, Cargnelutti Filho A & Pires FR (2007) Nutrient cycling in off-season cover crops on a Brazilian savanna soil. Pesquisa Agropecuária Brasileira, 42:1269-1276.
• Carpim LK, Assis RL, Braz AJBP, Silva GP, Pires FR, Pereira VC, Gomes GV & Silva AG (2008) Nutrient release from pearl millet in different phenological stages. Revista Brasileira de Ciência do Solo, 32:2813-2819.
• Crusciol CAC, Nascente AS, Borghi E, Soratto RP & Martins PO (2015) Improving soil fertility and crop yield in a tropical region with palisadegrass cover crops. Agronomy Journal, 107:2271-2280.
• Cunha EQ, Stone LF, Didonet AD, Ferreira EPB, Moreira JAA & Leandro WM (2011) Chemical attributes of soil under organic production as affected by cover crops and soil tillage. Revista Brasileira de Engenharia Agrícola e Ambiental, 15:1021-1029.
• Donagema GK, Campos DVB, Calderano SB & Teixeira WG (2011) Manual de métodos de análise de solo. 2ª ed. Rio de Janeiro, Embrapa Solos. 230p.
• Falleiros RM, Souza CM, Silva CSW, Sediyama CS, Silva AA & Fagundes JL (2003) Influence of tillage systems on the chemical and physical attributes of a soil. Revista Brasileira de Ciência do Solo, 27:1097-1104.
• FAO (2014) World reference base for soil resource 2014. International soil classification system for naming soils and creating legends for soil maps. Roma, FAO. 192p. (World Soil Resource Reports, n. 106).
• Faostat (2015) Production: Crops. Available at: <Available at: www.faostat.fao.org >. Accessed on September 29^th, 2015.
  » www.faostat.fao.org
• Filizadeh Y, Rezazadeh A & Younessi Z (2007) Effects of crop rotation and tillage depth on weed competition and yield of rice in the paddy fields of Northern Iran. Journal of Agricultural Science and Technology, 10:99-105.
• Franzluebbers AJ & Hons FM (1996) Soil-profile distribution of primary and secondary plant-available nutrients under conventional and no tillage. Soil & Tillage Research, 39:229-239.
• Garcia RA, Crusciol CAC, Calonego JC & Rosolem CA (2008) Potassium cycling in a corn-brachiaria cropping system. European Journal of Agronomy, 28:579-585.
• IIAM (2015) Da parceria IIAM/IRRI Moçambique terá um novo tipo de arroz. Available at: <Available at: http://www.iiam.gov.mz/index.php?option=com_content&view=article&id=154:mobiqueterm-novo-tipo-de-arroz&catid=8:timas&Itemid=28 >. Accessed on: October 26^th, 2015.
  » http://www.iiam.gov.mz/index.php?option=com_content&view=article&id=154:mobiqueterm-novo-tipo-de-arroz&catid=8:timas&Itemid=28
• IRRI (2015) Mozambique. Available at: < Available at: http://irri.org/our-work/locations/mozambique >. Accessed on: October 21^st, 2015.
  » http://irri.org/our-work/locations/mozambique
• Ivanic M & Martin W (2008) Implications of higher global food prices for poverty in low-income countries. Agricultural Economics, 9:405-416.
• 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 Development, 39:77-86.
• Mahmoudi M, Rahnemaie R, Soufizadeh S, Malakouti MJ & Eshaghi A (2011) Residual effect of thiobencarb and oxadiargyl on spinach and lettuce in rotation with rice. Journal of Agricultural Science and Technology, 13:785-794.
• Moreti D, Alves MC, Valerio Filho WV & Carvalho MP (2007) Soil chemical attributes of a red latosol under different systems of preparation, management, and covering plants. Revista Brasileira de Ciência do Solo, 31:167-175.
• Nascente AS, Crusciol CAC & Cobucci T (2013b) The no-tillage system and cover crops Alternatives to increase upland rice yields. European Journal of Agronomy, 45:124-131.
• Nascente AS, Crusciol CAC, Stone LF & Cobucci T (2013a) Upland rice yield as affected by previous summer crop rotation (soybean or upland rice) and glyphosate management on cover crops. Planta Daninha, 41:147-155.
• Nascente AS, Stone LF & Crusciol CAC (2015) Soil chemical properties affected by cover crops under no-tillage system. Revista Ceres, 62:401-409.
• Nayar NM (2014) Origins and Phylogeny of rice. Trivandrum, Elseiver. 324p.
• Nokkoul R & Wichitparp T (2014) Effect of drought condition on growth, yield and grain quality of upland rice. American Journal of Agricultural and Biology Science, 9:439-444.
• Oliveira AP, Silva JA, Lopes EB, Silva EE, Araújo LHA & Ribeiro VV (2009) Productive and economic yield of cowpea bean as a function of levels of potassium. Ciência and Agrotecnologia, 33:629-634.
• Oliveira TK, Carvalho GJ & Moraes RNS (2002) Cover crops and their effects on bean plant in no-tillage system. Pesquisa Agropecuária Brasileira, 37:1079-1087.
• Pacheco LP, Leandro WM, Machado PLOA, Assis RL, Cobucci T, Madari BE & Petter FA (2011) Biomass production and nutrient accumulation and release by cover crops in the off-season. Pesquisa Agropecuária Brasileira, 46:17-25.
• Planeta Arroz (2015) Até 2020 Moçambique prevê diminuir dependência de importação de arroz. Available at: < Available at: http://www.planetaarroz.com.br/site/noticias_detalhe.php?idNoticia=12995 >. Accessed on: October 21^st, 2015.
  » http://www.planetaarroz.com.br/site/noticias_detalhe.php?idNoticia=12995
• Ricepedia (2015) Mozambique. Availble at: < Availble at: http://ricepedia.org/index.php/mozambique >. Accessed on: October 21^st, 2015.
  » http://ricepedia.org/index.php/mozambique
• Rosolem CA, Werle R & Garcia RA (2010) Nitrogen washing from C3 and C4 cover grasses residues by rain. Revista Brasileira de Ciência do Solo, 34:1899-1905.
• Sá JCM (1993) Manejo da fertilidade do solo no plantio direto. Castro, Fundação ABC. 96p.
• Sanchez PA & Swaminathan MS (2005) Hunger in Africa: the link between unhealthy people and unhealthy soils. The Lancet, 365:442-444.
• SAS Institute Inc. (1999) Procedure guide for personal computers: Version 5. Cary, Statistical Analysis Systems Institute. CD-ROM.
• Torres JLR & Pereira MG (2008) Potassium dynamics in crop residues of cover plants in Cerrado. Revista Brasileira de Ciência do Solo, 32:1609-1618.
• Walkley A & Black IA (1934) An examination of the Degtjareff method for determining soil organic matter and a proposed modification of the chromic acid titration method. Soil Science, 37:29-38.
• Yahuza I (2011) Review of some methods of calculating intercrop efficiencies with particular reference to the estimates of intercrop benefits in wheat/faba bean system. International Journal of BioScience, 1:18-30.
• Yoshida S (1981) Fundamentals of rice crop science. Manila, International Rice Research Institute. 270p

Publication Dates

• Publication in this collection
  Nov-Dec 2017

History

• Received
  29 Feb 2016
• Accepted
  10 Nov 2017