Cover plants with potential use for crop ‐ livestock integrated systems in the Cerrado region

The objective of this work was to evaluate the effects of lignin, hemicellulose, and cellulose concentrations in the decomposition process of cover plant residues with potential use in no-tillage with corn, for crop-livestock integrated system, in the Cerrado region. The experiment was carried out at Embrapa Cerrados, in Planaltina, DF, Brazil in a split plot experimental design. The plots were represented by the plant species and the subplots by harvesting times, with three replicates. The cover plants Urochloa ruziziensis, Canavalia brasiliensis, Cajanus cajan, Pennisetum glaucum, Mucuna aterrima, Raphanus sativus, Sorghum bicolor were evaluated together with spontaneous plants in the fallow. Cover plants with lower lignin concentrations and, consequently, higher residue decomposition such as C. brasiliensis and U. ruziziensis promoted higher corn yield. High concentrations of lignin inhibit plant residue decomposition and this is favorable for the soil cover. Lower concentrations of lignin result in accelerated plant decomposition, more efficient nutrient cycling, and higher corn yield.


Introduction
The basic uses of cover crops include protecting the soil against erosion, increasing soil organic matter, improving soil quality, promoting carbon-sequestration, enabling producers to use less water and fertilizers and to control weeds.Cover crops are also more efficient at nutrient cycling and play an important role as a component of the no-tillage system (Machado & Silva, 2001;Diekow et al., 2005;Hollinger, 2005;Costa et al., 2008).In crop-livestock integrated systems, cover plants improve soil physicochemical and biological properties (Marchão et al., 2009), contribute to recover degraded pastures and areas, and have positive impacts on soil quality (Lal, 2002).Some cover plants, such as Cajanus cajan, Sorghum bicolor and Pennisetum glaucum, can improve the quality of animal feed.The selection of plant species for use as cover crops depends on the plant ability to adapt to climate and soil, as well as on the interest of the producer (Araújo & Almeida, 1993;Wutke et al., 2009).Maintaining crop residues on the soil surface protects the soil and enables a more Pesq.agropec.bras., Brasília, v.46, n.10, p.1200Brasília, v.46, n.10, p. -1205Brasília, v.46, n.10, p. , out. 2011 efficient nutrient cycling (Muhr et al., 1999;Pitol et al., 2006;Machado & Assis, 2010).
Biotic and abiotic factors, such as microorganisms, harvesting periods, plant flowering and maturation, air and soil temperatures, rainfall distribution, soil humidity, and the chemical composition of crop residues (C/N, lignin, hemicellulose, and cellulose) affect the biomass production, decomposition rates and the establishment of coverage on the soil surface (Araújo & Almeida, 1993;Aita & Giacomini, 2003;Sodré Filho et al., 2004;Carvalho et al., 2008Carvalho et al., , 2009)).
In the Cerrado region, some grasses such as P. glaucum are more cultivated as cover crops, due to their higher drought tolerance and biomass production, as well as the lower costs of their seeds (Lara Cabezas et al., 2004;Pacheco et al., 2011).Urochloa ruziziensis adapts easily and well to Cerrado's low fertility soils and climate, shows a high biomass production rate along the year and functions as an efficient soil cover.However, studies on its use as a cover crop are scarce.
Despite the advantages of using leguminous plants as soil coverage (Elfstrand et al., 2007;Wutke et al., 2009), which include the ability to use nitrogen from atmosphere, the presence of deep root systems and a high biomass production, the low C/N ratio of the tissues of leguminous plants favors the decomposition process and limits the establishment of soil coverage.However, plant species which show a higher C/N ratio, have lower plant residue decomposition rates, and consequently, are more efficient as soil coverage (Carvalho et al., 2008(Carvalho et al., , 2009)).
The objective of this study was to evaluate the effects of lignin, hemicellulose, and cellulose concentrations in the decomposition process of cover plant residues with potential use in no-tillage with corn, for crop-livestock integrated system, in the Cerrado region in Brazil.

Materials and Methods
The experiment was carried out at Embrapa Cerrados, Planaltina, DF (15º36'37"S, 47º44'37"W), in the Central West region of Brazil.Corn was cultivated in an area followed by cover plants planted at the end of the rainy season, in the first year, and this system was repeated during the second year.The soil of the region is classified as Typic Acrustox -Latossolo Vermelho, according to the Brazilian soil classification (Santos et al., 2006).The area is located within the morphoclimatic domain of Cerrado.The climate is Aw, tropical with rainy summer, according to the classification of Köppen.In this region, the average annual rainfall oscillates between 1,400 and 1,600 mm, and the average annual air temperature varies between 22 and 27ºC (Adámoli et al., 1987).
The The experimental design consisted of split plots, which were represented by plant species and subplots, which were represented by harvesting times (cut at flowering and at maturation), with three replicates.
The analyses of dry matter at 105ºC, acid-detergent fiber (ADF), neutral-detergent fiber (NDF), and lignin were performed with the sequential method (Robertson & Van Soest, 1981).The concentrations of hemicellulose and cellulose were determined by the differences between NDF and ADF, and between ADF and lignin, respectively.
Decomposition of plant residues was determined by using 20x20 cm nylon bags with 2x2 mm sized meshes containing 10 g of cut plant residues, which were oven-dried at 65°C for 72 hours.The nylon bags were removed from the experimental area at 30, 90, 150, 180 and 210 days after having been placed on the soil, and the material was weighed and oven-dried at 65°C for 72 hours once again.After this, the oven-dried material was put in porcelain crucibles and placed in a muffle furnace at 600°C for at least eight hours, in order to obtain the final inorganic material.
The decomposition rate, for each period, was calculated according to Santos & Whitford (1981).Based on the decomposition percentage, the rate of the remaining plant residues on the soil was determined as the difference between the total initial amount of residues (100%) and each of the decomposition rates.The data were adjusted to the simple linear regression model (Y = ax + b), with the rate of the remaining residues as a function of the accumulated rainfall, Pesq.agropec.bras., Brasília, v.46, n.10, p.1200Brasília, v.46, n.10, p. -1205Brasília, v.46, n.10, p. , out. 2011 for each evaluation period, and transformed into a base-10 logarithm.
Analysis of variance with repeated measures (ANOVA) was applied to evaluate the effects of plant species (plots), the effects of cover crops cut times (subplots) and the interaction between these two factors (SAS Institute, 1999).The test of multiple comparisons of averages (Tukey-Kramer at 5% probability) was applied to the treatments and to the interactions showing significance.

Results and Discussions
On average, cover crops cut at flowering had higher concentrations of hemicellulose and lignin.Cover crops cut at the end of their cycle (at maturation) showed higher concentrations of cellulose (Table 1).There was a significant interaction between cover plants and harvesting periods (Table 2).Cajanus cajan and M. aterrima had the lowest contents of lignin at maturation, probably due to the higher concentrations of protein and to the diluting effect of the nonfibrous carbohydrates in the grains.Pennisetum glaucum, however, had the highest concentration of lignin in the tissues at maturation, due to lower grain yield during the harvest period.Urochloa ruziziensis, C. brasiliensis, M. aterrima, R. sativus and natural fallow did not show differences in lignin concentration, in the plant tissue, at flowering and maturation.The maturation of the plants led to higher concentrations of cell wall fibers, lignification and cell wall thickening and, consequently, to a greater resistance to decomposition (Wagner & Wolf, 1999) (Table 3).
Cajanus cajan had higher concentrations of lignin at flowering (Table 2), which explains the lower decomposition rate of plant residues (Table 3) in comparison to the other studied plant species (Carvalho et al., 2008(Carvalho et al., , 2009)).A significantly higher concentration of lignin was also observed in the samples of M. aterrima, which were cut during flowering.When cut during maturation, P. glaucum had significantly higher plant tissue lignin concentrations, which explains the slower decomposition rate of plant residues at this stage of development.Despite its high hemicellulose concentrations, S. bicolor, had a slow decomposition rate due to its high C/N ratio.Urochloa ruziziensis and C. brasiliensis had significantly lower concentrations of lignin in plant tissue and faster residue decomposition rates, thus confirming the relation between lignin concentration and the decomposition process (Carvalho et al., 2008(Carvalho et al., , 2009)).
Cover plants with the lowest lignin concentrations and, consequently, faster residue decomposition rates, such as C. brasiliensis, U. ruziziensis and P. glaucum, contributed to higher corn yields (Table 4).However, cover plants with higher concentrations of lignin, such as C. cajan, S. bicolor and M. aterrima, had slower crop residue decomposition rates and promoted lower yields of corn planted in succession.The high quantity of Table 1.Effect of harvesting period on hemicellulose, cellulose and lignin concentrations (%) of cover plants (1) .Means followed by the same letters, lower case in the columns and upper case in the rows, do not differ by Tukey-Kramer test at 5% probability.
Pesq. agropec.bras., Brasília, v.46, n.10, p.1200-1205, out.2011 biomass, produced by C. brasiliensis, may also have favored the greater corn yield, because it accumulated large quantities of N in its aerial parts and this led to a higher release of nutrients, especially nitrogen (Muhr et al., 1999;Carvalho et al., 2008).Despite S. bicolor high biomass production (Table 4), its low decomposition rate of residues led to competition for nitrogen, and this decreased corn yield which had been planted in succession.
The data of lignin, hemicellulose and cellulose concentrations are complementary to the biomass production, which is an important information when choosing a cover plant for an integrated production system (Carvalho et al., 2008(Carvalho et al., , 2009;;Machado & Assis, 2010;Pacheco et al., 2011).Data on the quality of lignin, cellulose and hemicellulose and the quantity of biomass should support the producer decision-making process, as well as the choice regarding the best plant which will make up the no-till integrated production system (Machado & Silva, 2001).The use of C. brasiliensis and U. ruziziensis probably led to a more efficient nutrient cycling, and this increased corn yield in the crop-livestock integrated system (Carvalho et al., 2009;Marchão et al., 2009).In the case of an integrated system with soybean, which has a fast decomposition rate, cover plants with slower decomposition rates, such as C. cajan and S. bicolor, can be used in order to increase the carbon-sequestration of the soil (Silva et al., 1994;Hollinger et al., 2005;Costa et al., 2008).
The use of cover plants is essential for the sustainability of crop systems, especially in crop-livestock integrated systems on highly weathered soils and dry tropical climates as is found in the Cerrado region (Muhr et al., 1999;Machado & Silva, 2001;Lal, 2002, Pitol et al., 2006;Marchão et al., 2009;Machado & Assis, 2010).
Under the no-tillage system, C. cajan, S. bicolor, P. glaucum, and M. aterrima are suitable species for soil cover by plant residues as they fulfill an essential condition, due to their higher proportion of lignin and, consequently, lower decomposition rates (Carvalho et al., 2008(Carvalho et al., , 2009)).Nonetheless, for faster nutrient cycling, U. ruziziensis, C. brasiliensis, and R. sativus should be used because of their higher decomposition rates and, consequently, their contribution to increase the crop production in integrated systems.Table 4. Production of dry matter of cover plants and corn yield (1) .

Table 3 .
Decomposition process of plant residues harvested at flowering (A) and at maturation (B) in function of the accumulated rainfall quantity.
(1) Accumulated rainfall was transformed into a base-10 logarithm to obtain linear regression model.** and *Significant at 1 and 5% probability, respectively.
Means followed by the same letters, in the column, do not differ by Tukey-Kramer test, at 5% probability.