Production, decomposition of residues and yield of maize and soybeans grown on cover crops

The residues of cover plants and crops left on the soil surface can influence decomposition, nutrient cycling and follow crop yield. The objective was to evaluate the production of dry biomass (BD), of residues decomposition rate and yield of maize and soybean grown on different soil covers. The experimental design was a randomized block scheme banded, with four covers: Brachiaria, sun hemp, pearl millet and fallow period in autumn/winter, with maize and soybean crop in the spring/summer. We evaluated BD, decomposition through bags of decomposition and yield of maize and soybeans. Pearl millet and sunhemp were the covers which produced more of BD in the fall/winter period. the rate of decomposition of plant residues in maize and soybeans is slow in the dry period and accelerated in the rainy period; maize yield was higher when grown on sunhemp and brachiaria in the years evaluated, while for soybeans there were no differences on any measured coverage; maize and soybeans have grain yields higher than the regional average when grown on different soil covers.


INTRODUCTION
The sustainability of the tillage system (SSD) in the Brazilian Cerrado is linked to biomass production in quantities sufficient to provide adequate ground cover to compensate for the high rates of decomposition of plant residues occurring in these regions (CHIODEROLI et al., 2012). This biomass has been derived from the cultivation of plants of roofs and crop residues left on the soil surface after harvest cash crops.
The quantity of biomass produced in winter in the Cerrado has been lower compared to the one quantified in summer. However, some results show that brachiaria, sunnhemp, pearl millet, jack beans and fallow are adapted to the soil and climate of this biome, producing plant residues in high amount in both planting times (BOER et al., 2008;CARVALHO et al., 2011;CHIODEROLI et al., 2012;MENEZES et al., 2009;TEIXEIRA et al., 2012;TORRES;PEREIRA;FABIAN, 2008).
For annual crops (soybean, maize, sorghum, rice and beans) grown in succession to these covers, these have contributed to relative quantity of residues, however, Kluthcouski and Yokoyama (2003) state that these residues are unlikely to reach the quantity and longevity necessary for full protection of the soil surface.
Maize has been presented as a culture of great significance for the consolidation of the SSD in the Cerrado, with production crop residues above 6 Mg ha -1 (GUIMARÃES et al., 2006). This material has a low rate of decomposition, providing extended covers of the soil of nutrients and high cycling (SILVA et al., 2008). With respect to soybeans, biomass production is low, reaching 5.5 Mg ha -1 (PADOVAN et al., 2006), with high rate of decomposition of its residues (GONÇALVES et al., 2010).
Depending on the soil cover which preceded the cultivation of annual crops and edaphoclimatic conditions of the region, decomposition rates can vary considerably. In southern Brazil, Gonçalves et al. (2010) evaluated the decomposition of maize straw and estimated mass low of 40% and 65% after 150 and 360 days, respectively. In the Cerrado, Teixeira Neto (2002) found that after 180 days brachiaria straws intercropped with maize showed decomposition of 66% and 77% of sole maize. Kliemann, Braz and Silveira (2006) in this same consortium, found losses of 56% and 86% after 150 and 360 days, respectively. In a study with soybean, Gonçalves et al. (2010) estimated the decomposition of crop residues after 113 days in the field on the SSD by 33% and 51% in the conventional system. Padovan et al. (2006) found that the decomposition of soybean straw after 115 days in the field was of.
The maintenance of ground cover plants residues and those of commercial crops on the soil as well as their decomposition has been an important variable in the cycling of nutrients that contributes to the maintenance (CRUSCIOL; SORATTO, 2009;PACHECO et al., 2011), increased maize productivity (CARVALHO et al., 2004a;CARVALHO et al., 2011;TORRES et al., 2005) and soybean productivity (CARVALHO et al., 2004b;CHIODEROLI et al., 2012;TORRES;PEREIRA;FABIAN, 2008), however these values vary for each region and should be evaluated locally. This study aimed to evaluate the production of biomass, the decomposition rate of the waste and yield of maize and soybeans when grown on different cover crops in Cerrado miner. The soil of the area was classified as dystrophic Red Latosol (SANTOS et al., 2006), sandy clay loam texture, with plow layer (0.0 to 0.20 m), 210 g kg -1 clay, 710 g kg -1 sand and 80 g kg -1 silt, pH H 2 O 5.9; 15.2 mg dm -3 P (Mehlich-1); 2 mmol c dm -3 K; 12 mmol c dm -3 Ca 2+ ; 4.0 mmol c dm -3 Mg 2+ , 21 mmol c dm -3 H+Al and 19 g kg -1 organic carbon.

MATERIAL AND METHODS
The experimental design was a randomized, banded in the scheme, with four replications. The treatments used were four soil covers: brachiaria (Urochloa brizantha cv. Marandu), sunnhemp (Crotalarea juncea), pearl millet ADR 300 (Pennisetum glaucum L.) and fallow (spontaneous vegetation with a predominance of Poaceae) into sixteen plots of 180 m 2 (18 x 10 m), which after management (desiccation) were divided into areas of 90 m 2 (9 x 10 m) and seeded maize and soybean crop in succession.
The cover crops were grown in autumn/winter (April/July) period in 2007/08 and 2008/09. Sowing was mechanized in spacing of 0.45 m between rows and density of 50, 25 and 60 seeds per meter brachiaria, sunnhemp, pearl millet, respectively. The covers were desiccated in full bloom, applying a dose of 1440 g ha -1 of glyphosate + 600 g ha -1 of Paraquat. The evaluation of dry biomass (BD) was performed on an area of 2 m 2 per plot, was harvested where all vegetation, this material was then dried at 65 °C for 72 hours, weighed and the results expressed as Mg ha -1 . Production, decomposition of residues and yield of maize and soybeans grown on cover crops We used maize seeds of semi-early cycle from Pioneer, seeded with 0.90m between rows, with 55000 plants ha -1 . At sowing 400 kg ha -1 of the formula 05-15-10 + 0.5% Zn were applied. 100 kg ha -1 of N and 80 kg ha -1 K 2 O were applied at 20 and 40 days after sowing. For soybeans, the seed used was BRSMG of semi-early cycle, with 0.45m between rows, with 220000 plants ha -1 . 200 kg ha -1 of formula 0-20-15 + 2.5% Zn were applied at sowing.
To evaluate the decomposition the bags of decomposition (litter bags) were used (SANTOS; WHILFORD, 1981) with a 2 mm mesh opening, with 0.04 m 2 (0.20 x 0.20 m), were placed in 20 g of maize and soybean dry biomass. The 24 bags per plot were distributed in early May 2008 and 2009, and harvested four bags per plot at 15, 30, 60, 120 and 240 days after deployment in the field. These materials passed through pre-cleaning screens on and were dried in forced air at 65 °C until constant weight in order to determine the remaining biomass of waste.
In order to describe the decomposition of maize and soybean the mathematical model of the exponential type X = X o e -kt was applied, where X is the amount of dry biomass remaining after a period of time (t), in days; X o is the initial amount of dry biomass and k is the constant of the decomposition of the residue (THOMAS; -2007  467  232  100  106  25  0  21  0  40  154  266  441 1852  2008  468  362  300  244  67  5  0  21  42  95  146  430 2180  2009  335  338  246  61  44  1  12  11  58  170  156  325 1758 ASAKAWA, 1993). With the value of k, we calculated the half-life (T½ life) of the remainder dry biomass, using the formula T½ life = 0.693/k (PAUL; CLARK, 1996).
Mathematical equations and regression analyzes were developed with the help of SigmaPlot software, version 10. The results were analyzed for normality and homogeneity of data and subjected to analysis of variance and the averages compared by Tukey test at 5% of probability.

RESULTS AND DISCUSSION
Pearl millet and fallow were the covers that stood out significantly in 2007, which is higher when compared to the values produced of dry biomass in 2008 (Table 2). These values were similar to those reported by Torres, Pereira and Fabian (2008) and Fabian (2009) in the same experimental area. A consequence of higher values and better distribution of rainfall, which were recorded 106 and 25 mm in the month of April, 244 and 67 mm of rain in May in the years 2007 and 2008, respectively, higher values of production of dry biomass for all covers were expected, since these months in the study area are characterized by low rainfall (Table 1).
In studies conducted with the same cover, sowing period and the Cerrado (FABIAN, 2009;PACHECO et al., 2011;SODRÉ FILHO et al., 2004;TORRES;FABIAN;PEREIRA, 2008) values of dry biomass production were observed similar to those observed in this study. However, in some other areas of Cerrado, we have seen high yields of dry biomass, even with the covers sown in the beginning of the dry period (CARVALHO et al., 2004a;KLIEMANN;BRAZ;SILVEIRA, 2006), which are influenced by the local edaphoclimatic conditions.
In the production of dry biomass of maize and soybean on the cover plants there were no differences when analyzed separately the year (2007/08 and 2008/09), however the dry biomass production of maize in 2007/08 was higher on all roofs (Table 3). The production of dry biomass in maize and soybeans  is considered high and can be justified by the time the samples were conducted, because plants were in full flowering, with higher amount of green biomass. These values are above 6.0 Mg ha -1 obtained by Guimarães et al. (2006). Compared to soybean, there was no difference between the production of dry biomass of the crop on the covers. The values obtained were close to 5.5 Mg ha -1 of Padovan et al. (2006).
The rate of decomposition of crop residues on maize and soybeans covers the years 2007/08 and 2008/09 was evaluated through the remaining dry biomass, that occurred at an accelerated rate in the first 30 days and then slowly until completing 120 days (Figure 1). This behavior can be explained by the decrease in rainfall in the period from May to August (Table 1), with lower ambient temperature. After 240 days, the rate of decomposition increased with the return of precipitation and increasing temperature.  Analyzing the remaining residues within bags of decomposition was observed that the biomass losses for maize in 2007/08 were 22.8, 17.7, 20.3 and 15.0% at 120 days and increased to 33, 5, 23.6, 27.8 and 24.3% at 240 days after the distribution of decomposition bags. These losses for soybeans were 31.0, 23.8, 27.4 and 23.1 at 120 days and increased to 40.3, 36.4, 39.5 and 41.4% at 240 days, over the straw fallow, pearl millet, sunnhemp and brachiaria, respectively.
In 2008/09 the biomass losses for maize were 24.8, 27.5, 30.8 and 29.8% at 120 days and increased to 38.7, 47.4, 39.1 and 37.2% at 240 days. For soybeans, the values were 40.1, 38.5, 42.8 and 42.7 at 120 days and increased 63.2, 53.0, 53.7 and 57.3% at 240 days for fallow, pearl pearl millet, sunnhemp and brachiaria, respectively. The highest values of mass loss observed in this second year are due to the higher volume of rainfall recorded during the period (Table 1), which caused the increase of the rate of decomposition of residues. Production, decomposition of residues and yield of maize and soybeans grown on cover crops Some studies on the decomposition of the straw of maize and soybeans in the southern have lower values of biomass loss, to those observed in this study. In Carambeí-PR, Wisniewski and Holtz (1997) found that at 23, 40, 65, 89, 103 and 149 days after handling the mass losses of maize straw were of 31, 27, 59, 23, 23 and 49%, respectively. In Londrina, Gonçalves et al. (2010) estimated the loss of biomass of maize in areas under SSD of 40 and 67% at 150 and 360 days after the handling, respectively, whereas for the soybean the biomass losses were of 33% at 113 days.
These values are justified by the different edaphoclimatic conditions that occur in this region, especially the lowest temperatures, which cause a reduction in the rate of decomposition of residues.
In the Cerrado of Goiás, Kliemann, Braz and Silveira (2006) evaluating the consortium of maize and brachiaria observed biomass losses of maize of 56% at 150 days. In Seropédica-RJ, Padovan et al. (2006) evaluated the biomass losses for soybeans at 55, 85 and 115 days after the emergence and found 51, 56 and 40% of the remaining dry biomass. These regions have high temperature and rainfall. These conditions are identical to those observed in this study, thus the values of biomass losses are also similar.
Analyzing the half-life (T ½ life) of maize straw and soybean in 2007/08 and 2008/09, it is observed that the smaller values occurred when cultures were grown on sunnhemp (Table 4). In these areas there was a more rapid decomposition of crop residues due to the larger amount of N available in the soil through biological fixation of sunhemp (BOER et al., 2008;CARVALHO et al., 2011;MENEZES et al., 2009;PACHECO et al., 2011;TORRES et al., 2005) thereby increasing the activity of decomposing organisms of organic matter occurs. Gonçalves et al. (2010) emphasize that together with the carbon assimilation, microbial population needs to assimilate N for the synthesis of many compounds, therefore the lower the C/N ratio of the vegetable residue the greater the rate of decomposition.
This pattern can be observed in Table 4, where the highest values of T ½ life of maize straw and soybean occurred on brachiaria and pearl millet in 2007/08, pearl millet and fallow in 2008/09, areas in which a lower availability of N occurs, once they are covers that have higher C / N ratio when compared to sunhemp, as verified by Torres et al. (2005) and Carvalho et al. (2011).
The C/N ratio of the maize and soybeans straw also influences the residues decomposition, as in maize this relation ranges from 60:1 to 69:1 (SILVA et al., 2008), whereas in soybean, values are smaller than 30:1 (GONÇALVES et al., 2010). Although not determined in this study, the C/N ratio of these cultures appeared to influence the rate of decomposition because the life T ½ of the maize straw was in most cases larger when compared to soybeans.
The values obtained for the decomposition rate and rainfall showed positive correlations. (Table 5). These correlations show that as rainfall increased (Table 1), the greater were the rates of decomposition and the biomass losses of crop residues of maize and soybeans during the evaluation period.
Various studies have were conducted in Cerrado that proved the influence of rainfall on the rate of decomposition of plant residues of different ground covers (BOER et al., 2008;CARVALHO et al., 2011;MENEZES et al., 2009;PACHECO et al., 2011;    In this study it was observed that the decomposition of maize straw occurs slowly and for soybeans at an accelerated rate, the values increase gradually with increasing precipitation, as observed by other authors (GONÇALVES et al., 2010;GONÇALVES;SARAIVA;KLIEMANN;BRAZ;SILVEIRA, 2006;PADOVAN et al., 2006;SILVA et al., 2008). Furthermore, it was found that this decomposition is influenced by the predecessor cultures, for smaller values of T ½ life of maize straw and soybean occurred on the sunhemp straw because the high fixation capacity of atmospheric N 2 of this culture offers greater amount of N in the soil after being managed, causing increased activity of decomposing organisms (GONÇALVES et al., 2010).
The use of grasses as cover crops succeeding other grassy main culture can promote a low N availability in the soil and affect grain yield. This was confirmed in 2007/08 as the highest grain yield maize was observed when the culture was grown on sunnhemp in 2007/08, but was not repeated in 2008/09 because the highest yield occurred on brachiaria. For soybeans there was no difference between the covers in those years ( Table 6). The highest yields observed for maize in 2007/08 compared to 2008/09 are justified by better distribution and higher rainfall occurred this year, while the same was not true for soybeans. Production, decomposition of residues and yield of maize and soybeans grown on cover crops Table 5  Grain yield for recorded for maize and soybeans in the two years evaluated is high for the region, which has estimated averages of 6.0 Mg ha -1 for maize and 3.0 Mg ha -1 for soybean (COMPANHIA NACIONAL DE ABASTECIMENTO, 2011). Some previous studies were conducted in the same experimental area during 2000/01 and 2001/02, showed average grain yield maize from 5.4 Mg ha -1 (TORRES et al., 2005), in 2004/05, 2005 / 06 and 2006/07 of 6.9 Mg ha -1 by Fabian (2009). For soybeans, the yields observed in this study (Table 6) were similar to those observed by Torres et al. (2005) in 2000/01 and Fabian (2009Fabian ( ) in 2004Fabian ( /05, 2005.0 and 4.6 Mg ha -1 soybean on the same cover. The results of this study show that is taking place to improve the quality of the soil with the use of cover crops preceding the annual crop.

CONCLUSION
1. Sunhemp and pearl millet were the covers that produced more dry biomass during the autumn/ winter period; 2. The dry biomass production of maize and soybeans was not influenced by the soil cover evaluated; 3. The rate of decomposition of plant residues in maize and soybeans is slow in the dry period accelerated in the rainy period; 4. Maize yield was higher when grown on when grown on sunhemp and brachiaria in the years evaluated, while for soybeans there were no differences on any cover evaluated. 5. Maize and soybeans present grain yields higher than the regional average when grown on different soil cover.