No-tillage effect on carbon and microbiological attributes in corn grown in Manaus-AM , Brazil

The soil is the most important carbon reservoir, but agricultural practices involved in tillage systems decrease the soil carbon stock. This study aims to evaluate the effect of no-tillage (NT) on carbon and microbiological attributes of the soil in corn plantations in Manaus-AM, Brazil. Soil was sampled in a secondary forest (SF) and in corn grown under different tillage systems, including conventional tillage (CT), which makes use of plowing and harrowing operations, and no-tillage (NT). Soil variables studied included, carbon content, soil microbial carbon and basal respiration, metabolic and microbial quotients. The data were analyzed by analysis of variance and Tukey's test. The results demonstrated that NT resulted higher carbon, lower basal respiration, and lower metabolic quotient than CT and SF. However, soil microbial carbon was similar in all tillage systems studied. We conclude that NT corn cropping increased soil carbon content more than CT, while it decreased basal respiration and metabolic quotient when compared to NT and SF. Moreover, soil microbial biomass in corn was similar in all tillage systems studied. This research demonstrated the importance of NT to soil carbon conservation and soil management in corn cropping.


INTRODUCTION
Soil quality is very important for agricultural sustainability and environmental conservation (Khaledian et al., 2016).However, this quality is altered by interferences with the chemical, physical and biological properties of the soil (Cattelan et al., 1990;Andrist-Rangel et al., 2007;Cerdà et al., 2017) caused by the soil management practices *Corresponding author.E-mail: aleksander.muniz@embrapa.br.
Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License involved in the various tillage systems in use (Treseder, 2008;Lu et al., 2011).Differences in tillage systems are related to soil mobilization and disposal of plant residues (Lisboa et al., 2012).
Conventional tillage (CT) uses plowing operations that change soil structure and reduce soil organic matter and microbial activity (Allaiume et al., 2014).In contrast, notillage (NT) minimizes soil disturbance, thereby promoting the increase of carbon and microbial carbon in the soil (Bayer et al., 1997;Pérez et al., 2004).Microbial change is determined by microbial biomass, respiration and metabolic quotients (Tótola and Chaer, 2002;Gajda et al., 2012).
Microbial biomass is the live and active part of the soil organic matter (Tótola and Chaer, 2002).It is also a more evaluable organic matter reservoir (Roscoe et al., 2006).In addition, soil respiration can reflect disturbance and ecosystem productivity (Islam and Weil, 2000); while the microbial quotient was used in studies of soil organic matter dynamics (Tótola and Chaer, 2002).Moreover, tillage affects CO 2 liberation and soil compaction, which may enhance climate change and soil erosion (Bogunovic et al., 2017).
Effects of soil preparation on agricultural crops have scarcely been studied in the state of Amazonas.Thus, the objective of this work is to evaluate the effect of NT on carbon and microbiological attributes of the soil in corn grown in Manaus-AM.

MATERIALS AND METHODS
Soil sampling was carried out in March of 2012 in Ferrosols (WRB, 2014) located in Manaus (2°53'47.27"S,59°58'29.76"O)Amazonas.Spacing between samples was 10 m and sampling depth was 10 cm.Samples were obtained in secondary forestland and in a corn plantation under either CT with plowing and harrowing operations or under a NT, without soil movement.Ferrosols represent 18.15% of the soils in the state of Amazonas and correspond to 285,041.75 km 2 (Teixeira et al., 2010).These sampling areas are shown in Figure 1.Cultivation of corn under NT was carried out between 2008 and 2012.The crop was fertilized with 425 kg ha -1 of 4N-14P-8K at sowing.Additionally, 2 kg ha -1 Zn and 90 kg ha -1 N were applied by broadcasting.Finally, 1,500 kg.ha -1 dolomitic limestone was applied (CFSEMG, 1999).
Soil samples were sieved through a 2 mm mesh prior to chemical analysis according to Claessen et al. (1997).Results are summarized in Table 1.After that, the soil carbon was determined by wet oxidation method (Walkley and Black, 1934).Microbial biomass carbon (MBC) and basal respiration (BR) were evaluated by Infra-Red Gas Analyzer (IRGA) measurements according to Anderson and Domsh (1978).MBC was calculated using the formula below: Where: Other formulas were used for metabolic and microbial quotient respectively: The results were subjected to analysis of variance to detect significant effects (Fisher, 1925), while means were compared by Tukey's test (Pimentel-Gomes, 2009).The analysis was done using Matlab software using p<0.05 for Ct, RB and qCO2, and p<0.10 for CBM and qMic (Table 1).

RESULTS AND DISCUSSION
NT corn and SF showed higher soil carbon concentration  than conventional tillage (CT) corn (Table 2).This higher carbon concentration was due to plant residue accumulation during crop growth (Freixo et al., 2002;Lovato et al., 2004;Sisti et al., 2004;Hickmann and Costa, 2012;Asmann et al., 2014).Furthermore, it could be due to a decrease in mineralization rate (DICK, 1983).Conversely, carbon concentration in SF occurred because of rhizodeposition, fall of leaves and lixiviation (Richter et al., 1999).Rhizodeposition in the forest was affected by plant composition and litter quality (Wilcke and Lilienfein 2004;Wiesmeier et al., 2009).On the other hand, litter quality on the surface of the soil changed due to a high CN ratio and low biodegradability (Tejada et al., 2009) (Table 2).
Basal respiration was lower in NT corn than in the other systems studied (Table 2).These results differed from those observed by D' Andréa et al. (2002) in Goiás, Brazil, where no basal respiration differences between tillage systems for corn production were found.Moreover, the higher basal respiration in the CT corn crop, and SF was due to increased organic matter and nutrients available in the soil (Emmerling et al., 2000).
Microbial biomass carbon (MBC), was higher in SF than in CT corn, but showed no difference with respect to NT corn (Table 2).Higher MBC in SF is related to the higher quantity and quality of plant residues and organic matter quantity (Jacinthe et al., 2000;Fierer et al., 2009;Diacono and Montemurro, 2010).We recorded a decrease in MBC in CT corn, caused by the combined effect of a reduction in carbon input and a decrease in soil carbon stock (Kallenbach and Grandy, 2011).This decrease was also a result of agricultural practices, such as soil plowing and fertilizer use (Treseder, 2008;Lu et al., 2011).Results for NT in this study were different from those obtained in other studies, which showed higher CBM content in NT than in CT corn crops in Paraná state and the Cerrado areas of Brazil (Balota et al., 1998;Balota et al., 2004).However, CBM under NT showed a similar behavior to that observed in Minas Gerais state (Rangel and Silva, 2007).
The qCO 2 was lower for NT than for the other systems (Table 2).The same was observed in other studies on NT in southern Brazil (Balota et al., 2004).The higher qCO 2 in Ct corn may be due to the incorporation of plant residues into the soil (Ocio and Brookes, 1990).Therefore, the differences between NT and CT corn were due to microbial access to the substrate (Alvarez et al., 1995).Soil qCO 2 in CT corn showed an imbalance and was probably dominated by organisms growing faster (Sakamoto and Obo, 1994).Instead, the lower qCO 2 in NT corn and SF occurred due a greater efficiency of the microbial biomass in using environmental resources with lower carbon losses, such as CO 2 (Anderson and Domsch, 1985).Hence, the qCO 2 values in SF and NT corn indicated a greater stability of the microbial biomass in these systems (Santos et al., 2005).The higher qCO 2 value in SF than in NT likely occurred due to the more advanced stage of plant succession (Chapman et al., 2003).
Higher qMic was observed in SF than in CT corn, but it did not differ from qMic registered for NT corn (Table 2).Overall, qMic values measured in this study were higher than values measured in other areas of the Amazon rainforest, where qMic values ranged from 1.1% to 3.7% (Pfenning et al., 1992, Geraldes et al., 1995, Moreira and Malavolta, 2004).
Corn growing areas showed qMic values similar to those found in other Amazonian sites with 2.6% (Pfenning et al., 1992).Differences observed in qMic values in this study may have occurred due to the formation and conversion efficiency of recalcitrant organic matter into microbial biomass carbon (Sparling, 1992).We also observed that qMic values obtained in SF and in corn growing areas were 2.2% higher, which is the suggested value for a balanced soil (Jenkinson and Ladd, 1981).

Conclusions
The no-tillage corn system increased soil carbon content over that registered under the conventional tillage system.Moreover, no-tillage displayed lower basal respiration and metabolic quotient values, compared to conventional tillage and secondary forest.Furthermore, it showed similar microbial biomass to that which characterized these other systems.

Figure 1 .
Figure 1.Sampling areas in a corn plantation.A No-tillage, B conventional tillage and C secondary forest.
B*Means with the same letter within columns do not differ significantly by Tukey's test (* p <0.05; ** p <0.10).