Biomass and nutrient accumulation of leguminous green manure species in the Jaguaribe-Apodi region , Ceará , Brazil

The ability of leguminous green manure to incorporate significant amounts of nutrientes into the soil makes them excellent alternatives for the region agriculture systems. This study aimed to evaluate the potential use of six leguminous species (Crotalaria juncea, Canavalia ensiformes, Cajanus cajan, Crotalaria spectabilis, Dolichos lab lab and Mucuna deeringiana) as green manure in an area of the Jaguaribe-Apodi agricultural region, CE, Brazil. The experiments were carried out under field conditions in a randomized block design with five replications. Sampling of biomass of shoots was carried out 78 days after sowing on each plot. It was determined the total biomass and nutrients accumulation in the plants that grew in the plots (leguminous and weeds). We concluded that the use of Crotalaria juncea showed up as the most promising treatment in terms of biomass production and nutrient accumulation.


INTRODUCTION
The Baixo Jaguaribe region in the state of Ceará, is one of the areas with the greatest potential for irrigated agriculture in the Brazilian Northeast. The agricultural systems in the region use large amounts of inputs, especially mineral fertilizers, which can represent a large proportion of production costs. Thus, the use of leguminous green manure can be a alternative for recycle and incorporate significant amounts of nutrients, particularly the nitrogen (N), through fixation of atmospheric N, increasing economic returns and decreases the negative environmental impact (LIM, 2012).
There is wide variation in the yield of biomass and nutrient accumulation by leguminous species, which must be related to the intrinsic characteristics of each specie, the management of crop residues and soil and climate conditions prevailing. Variations in biomass production and accumulation of nutrients by green manure legumes in different environmental conditions were reported by Odhiambo (2011) and Thomas and Palaniappan (2012).
By evaluating the potencial biomass of leguminous species, we can indicate the best species for local conditions. In this respect, Crotalaria genus has excelled in studies of potential of green manure (MATTAR et al., 2015). The capacity to nutrients accumulation can also be a good criterion in the selection of the species to be used as green manure. In Sete Lagoas, MG, Brazil, Favero et al. (2000) evaluated the potential of five leguminous species, which have accumulated between 109 and 222 kg ha -1 of N.
Thus, the objective of this study was to evaluate the potential use of leguminous species as green manure in the Jaguaribe-Apodi region.
We performed the mechanical preparation of the area for seeding of legumes. No fertilizer was performed Sowing was performed manually on 02/10/2008, in rows spaced 0.50 m, at a depth of 2 to 3 cm and sowing densities of: 1.25, 0.50, 1.25, 5.00, 2.00 to 4.00 g m -1 for treatments L1, L2, L3, L4, L5 and L6, respectively.
It was installed micro sprinkler system in the area to supplement rainwater. The precipitation in the experimental period was 53.6 mm (February), 294.4 (March) and 138.8 (April).
Sampling of biomass of shoots was carried out 78 days after sowing (04/28/2008), using a square frame with an area of 0.5 m² on each plot. Leguminous plants and weeds were cut at ground level and weighed to determine biomass fresh weight.
At each sampling, part the material was dried in an oven with forced ventilation of air at 65 o C for 72 h and the mass of the plant material was determined. Then it was milled for the determination of carbon and nutrients. Carbon contents were determined by the Walkley-Black method described by Yeomans and Bremner (1988). Nitrogen was determined after sulfuric digestion by micro Kjeldahl method, and the macronutrients (P, K, Ca, and Mg) were determined after mineralization nitro-perchloric mixture as described by Silva (2009). The amount of original nutrient was estimed through product of its tenor and dried mass.
Data were subjected to analysis of variance and means were compared by Tukey test at 5 % probability. The significance of the coefficients of correlation and determination were assessed by t test at 5 % probability.

RESULTS E DISCUSSION
Treatment L1 (Crotalaria juncea) was not statistically different from L2 (Canavalia ensiformes) and L5 (lab-lab) for the production of fresh matter, however the L1 was superior to the others for dry matter production (Table 1). In fact, several studies have confirmed the potential of Crotalaria. juncea for dry matter production (CÁCERES; ALCARDE, 1995;PERIN, 2005;TORRES et al., 2005;TEODORO et al., 2011). According to Mattar et al. (2015), the Crotalaria genus stand out when studying green manure.
The dry mass yield of treatment L5 (Dolichos lab lab) was lower than the results presented by Faria (2004) for legumes in the Submédio São Francisco region of 10.64 and 12.69 Mg ha -1 with the mowing season at 128 and 175 days after planting, respectively (full flowering). For the other legumes, in general, the results were higher than those presented by Faria (2004), with emphasis on Crotalaria juncea, for which he obtained a dry mass yield of 5.71 Mg ha -1 (68 days after planting), while in this study it was 12.82 Mg ha -1 ( Table 1).
Some of these differences can be explained by the different cutting times. The availability of light and temperature in the Baixo Jaguaribe region may also have favored legumes in this work. Thomas and Palaniappan (2012) pointed to the sensitivity of some leguminous green manure to photoperiod, including Crotalaria juncea. The authors observed that the shorter growth period and low temperature restricted vegetative growth.
Cáceres and Alcarde (1995) also obtained biomass yields lower than those of the present experiment, except Crotalaria spectabilis, which they observed yields of 21.6 and 4.2 t ha -1 of green and dry matter, respectively. Cavalcante et al. (2015) also observed higher values of fresh mass (26.5 t ha-1) and dry mass (6.0 t ha-1) for this leguminous specie in an experiment conducted in Araíraca, Agreste of Alagoas. That can be explained by the delay in germination observed in all plots with Crotalaria spectabilis in this work.
The weeds may have influenced the biomass production of legumes, mainly because they have been grown in the rainy season, when competition with them is more intense, but not necessarily affect the total biomass yield. Favero et al. (2000) found no influence of weeds on yield of shoot dry weight, demonstrating that the presence of weeds did not reduce the total biomass yield.
We observed differences of nutrient accumulation in shoot biomass of legumes (Table 2). Legumes have accumulated on average almost 200 kg ha -1 N in shoots (Table 2). Table 1 -Fresh and dry mass, and percentage of dry matter in the biomass of legume treatments, at 92 days after planting.
Means followed by the same letter in the column do not differ by Tukey test at 5 % probability. (1) L1sunn hemp (Crotolaria juncea); L2jack bean (Canavalia ensiformis); L3dwarf pigeon pea (Cajanus cajan); L4showy crotalaria (Crotalaria spectabilis); L5lab-lab (Dolichos lab lab) and L6 -dwarf mucuna (Mucuna deeringiana). In general, the treatments produced high accumulations of N in biomass, which makes it interesting to use these species as a source of nitrogen in agroecosystems in the region studied. Treatment L2 presented the lowest C/N ratio, but it was significantly different only from treatments L4 and L5 (Table 2). These treatments still had low C/N ratio (<20), a parameter that can be decisive to speed mineralization of organic N in the soil.
However, it should be considered that the C/N ratio may be a characteristic quite variable within a farming system, having close relationship with the conditions for plant growth. Torres et al. (2005) found variations in C/N ratio of legumes used as cover crops in two growing seasons. That variation was explained by possible disfavoring of N fixation by legumes. According to Ribeiro Jr. and Ramos (2006), any factor that affects plant growth may influence the biological fixation of N.
In that sense, the successive planting of legumes may encourage the symbiotic fixation, increasing the inoculum potential of the soil, but also because of the influence of green manure in stimulating microflora and improving soil quality (BUZINARO et al., 2009).
Thus, smaller values of the C/N ratio should be associated with higher amounts of N in biomass, thereby suggesting potential for the rapid mineralization of residues of these species (RIBEIRO JUNIOR;RAMOS, 2006).
Overall, nutrients accumulated in biomass of legume species can be considered high and are comparable or superior to those found by Cáceres and Alcarde (1995), at 110 days after planting, except for N and Ca, for Crotalaria spectabilis, for which the same authors found amounts of 113.4 and 63 kg ha -1 , respectively. The accumulation of N in L1 (Crotalaria juncea) was superior to those presented by other authors such as Torres et al. (2005) Cavalcante et al. (2015) (138.5 kg ha -1 ), probably due to the high yield of dry mass obtained in this work. However, Teodoro et al. (2011) found accumulations of nutrients, in general, higher even for the N in Crotalaria juncea (514 kg ha -1 ) but with a cycle of 163 days (flowering).
Treatment L1 (C. juncea) showed higher accumulation of nutrients, except Ca, for which treatment L2 (jack bean) showed higher accumulation (Table 2). In general, one can say that C. juncea stood out as the most promising legume in terms of supply of biomass and nutrients, while the residues of jack bean showed lower C/N ratio, but showed lower biomass production in relation to L1, despite the competitive capacity of the jack bean against weeds due to its allelopathic potential (FARIA, 2004). The accumulation of P in C. juncea was similar to lablab (Table 2), although it has presented a dry mass production 2.7 times higher than the lab-lab. This can be explained by the high absorption capacity of P for this species.
There were no significant differences between treatments for the accumulation of K (Table 2), which may be explained by the high level of this nutrient in the soil.
The high accumulation of K presented by legumes can be very important when one wants to use those species associated with fruit trees, avoiding the loss of that nutrient through leaching, allowing its recycling.
In general, the characteristics evaluated showed that legume species showed a high accumulation of nutrients, demonstrating their potential for nutrient cycling in the agricultural system.

CONCLUSIONS
In general, the characteristics evaluated showed that legume species showed a high accumulation of nutrients, demonstrating their potential for nutrient cycling in the agricultural system.
Crotolaria juncea was the most promising specie in terms of biomass production and nutrient accumulation.