Emission of CO 2 and soil microbial activity in sugarcane management systems

School of Agricultural Engineering, University of Campinas, Av. Cândido Rondon, 501, Barão Geraldo, CEP 13083-875 Campinas, São Paulo, Brazil. Department of Exact Sciences, School of Agricultural and Veterinary Sciences of Jaboticabal, São Paulo State University, s/n, Santa Luzia CEP 14884900 Jaboticabal, São Paulo, Brazil. Department of Soils, Federal Institute of the Triângulo Mineiro, Uberaba Campus, Rua João Batista Ribeiro, 4000, CEP: 38064-790, Uberaba, State of Minas Gerais, Brazil. Department of Agronomy, Federal University Amazonas. Rua 29 de Agosto, 786, Centro, CEP.: 69.800-000, State of Amazonas, Brazil.


INTRODUCTION
The cycle of sugarcane cultivation has been the subject of studies because of the impacts caused in the soil and atmosphere, related mainly to the sugarcane burning system, which is a common practice in Brazil and whose main objective is to facilitate the manual cutting.To replace the burning system, the green sugarcane system was implemented, in which sugarcane is harvested mechanically, without prior burning, and the waste is deposited on the soil, on an average of 10 to 30 Mg ha -1 of mulch (Souza et al., 2005), which benefits the soil (Mendonza et al., 2000;Souza et al., 2005) and provides a favorable microclimate environment for the development of biological communities that will act in the subsequent process of decomposition of organic residues (Xu and Qi, 2001;Franchini et al., 2007;Matias et al., 2009).
The green sugarcane system can function as carbon mitigation for the environment, because despite the of CO 2 release during the microbial decomposition of the carbon that would be lost if the sugarcane was burned is incorporated into the soil (Panosso et al., 2008).The CO 2 release is due to the action of microorganisms in the process of decomposition of organic matter stimulated by the greater amount of substrate that the green sugarcane system provides.These organisms are widely used as indicators of soil quality as they are very sensitive to changes in soil management in the short time (Galdos et al., 2009), since physical or chemical attributes are not always sufficient to explain the variations that occur in the soil from the actions of its use and handling.
The influence of temperature and soil moisture on CO 2 emissions has already been reported in several studies (Kosugi et al., 2007;Panosso et al., 2008;Siqueira Neto et al., 2011;Lenka and Lal, 2013;Song et al., 2013); thus, research studies for new factors that have a relationship with CO 2 are of the utmost importance to try to understand the dynamics of this gas in the soil, as well as its stabilization and carbon buildup.Soil microorganisms can clarify many questions on CO 2 emissions, as the microbial activity is primarily responsible for the decomposition of organic residues, nutrient recycling and energy flow in the soil, this way exerting influence on the carbon storage, availability of nutrients for plants and CO 2 emissions (Jenkinson and Ladd, 1981).
The carbon of the microbial biomass and the basal respiration are the most used attributes in studies on the biological indicators of the soil, and microbial biomass is the most active living part of the soil organic matter, formed mainly by fungi and bacteria (Kaschuk et al., 2009), while soil respiration indicates the degree of activity of the biomass.These attributes are considered easy indicators of soil quality because of their high sensitivity to changes in management or climate; however, some attributes show difficulties when being interpreted if evaluated individually (Lopes et al., 2013), such as, for example the basal respiration, since high respiration values do not always indicate desirable conditions in the short term, as a high respiration rate can mean release of nutrients into the soil, and in the long term, loss of organic carbon to the atmosphere.
The short-and long-term temporal monitoring aids in the interpretation of the soil microbiological quality; moreover, Anderson and Domsch (1990) proposed relationships between the attributes aiming at a more interpretative approach and to establish dynamic relationships between biomass and microbial activity, such as metabolic quotient (qCO 2 ), which calculates the release of CO 2 per unit of biomass for a certain time, and microbial quotient (qMIC), which evaluates the availability of organic carbon for microbial activity.Thus, an ecosystem out of balance will present high values of qCO 2 and low values of qMIC, which indicates greater energy consumption and higher level of stress of the biomass (Anderson and Domsch, 1990;Evangelista et al., 2013;Kuwano et al., 2014).
Biological indicators of the soil have been used in studies on the efficiency of management systems, such as conventional and no-tillage (Martinéz et al., 2013;Alves et al., 2011), of different ecosystems, such as pasture and shrubs (Loureiro et al., 2010), of forestry and agroforestry systems (Silva et al., 2012) and on the comparison of the management systems of green sugarcane and burned sugarcane (Mendonza et al., 2000).In these studies, the soil microbial activity showed sensitivity to management efforts, thus strengthening the use of these attributes in the understanding of the stability of soil carbon.This way, this study aimed to evaluate the soil CO 2 flux and the soil microbial activity in the systems of burned sugarcane and green sugarcane.

MATERIALS AND METHODS
The study was conducted in the northeast of the State of São Paulo, near the coordinates 21°19'8" South and 48°7'24" West.The climate in the region is classified as B 2 rB'4a' by the Thornthwaite climate classification criterion.The soil of the area was classified as eutroferric Oxisol, clayey texture, with flat and undulating topography.
The areas evaluated were implanted in three sugarcane management systems: burned sugarcane (BS), green sugarcane implemented for five years (GS-5), one cycle with this system, and green sugarcane implemented for ten years (GS-10), two cycles with this system, being that both green sugarcane areas were converted from the scenario of prior burning before harvest.
We performed a chemical characterization of the soil in the areas before the start of the evaluations (Table 1), whose data showed increased base saturation (SB) in BS, which may be related to the large amount of ash deposited on the soil surface from the prior burning of the sugarcane, thus contributing with the immediate addition of mineral nutrients, such as K, Ca and Mg.
The burned sugarcane area was managed with the burning system since the '80s and in 2011-2012.The area of green sugarcane implemented for five years began to be harvested by the mechanized system from 2006, and the system of mechanized harvesting began in 2001 in the area of green sugarcane implemented for ten years.On the reform of the sugarcane plantation, which occurred in the areas of burned sugarcane (every 6 ratoons) and green sugarcane for ten years (in 2007), there was the mechanical elimination of the ratoon of the previous crop and subsoiling at the depth of 0.45 m in the planting furrows.Soon after, 2 t ha -1 of dolomitic limestone were applied.For the planting fertilization, 480 kg ha -1 of NPK in the 10-25-20 formulation were used.Over the years, on average, 100 m -3 ha -1 of vinasse and 300 kg ha -1 of urea or 200 kg ha -1 of ammonium nitrate were applied in the areas.
The evaluations of the soil CO 2 flux (FCO 2 ) and soil collection at the 0.00 to 0.10 m layer for biological analyses were performed in 30 points on a sampling mesh at the regular intervals of 1, 2 and 10 m (1 ha), whose points were georeferenced with the aid of a total station (model TC 305 Leica ® ) and DGPS (L1/L2 Hiper Lite Plus) (Figure 1).The evaluation of CO 2 was performed simultaneously in the three areas of study in the dry period of 2011 and wet period of Table 1.Chemical characterization of the soil in the management areas of burned sugarcane, green sugarcane for five years and green sugarcane for ten years, in Pradópolis, São Paulo, Brazil, 2011-2012.

Chemical attributes
Burned sugarcane Green sugarcane for five years Green sugarcane for ten years OM (1) (g kg (1) OM = organic matter; (2) SB = sum of bases; (3) CEC = cation exchange capacity; (4) V% = base saturation.2012 in the mornings (07:00 to 10:00 am).The evaluation was conducted with the aid of ground chambers, of the model LI-8100 (LICOR).The equipment is a closed system with internal volume of 991 cm 3 , with contact area with the soil of 71.6 cm 2 and placed on PVC collars previously inserted into the soil at a depth of 3 cm.Soil moisture was measured simultaneously with the measurement of the CO 2 concentration with the aid of a portable TDR (Campbell®).
The CO 2 emission by basal respiration in laboratory was evaluated for comparative purposes with the CO 2 emission evaluated in field, named in this paper as soil CO 2 flux (FCO 2 ), and the main difference is that the basal respiration calculates the CO 2 from soil microorganisms, whereas the FCO 2 calculates the CO2 from microorganisms and roots.
Sampling for microbiological analysis was performed on the 0.00-0.10m layer in two days of collection in each period (winter and summer) and area, amounting to 360 samples, which were kept under cooling until the analyses within a maximum of 30 days.The analysis for the carbon of the microbial biomass was carried out according to the fumigation-extraction method proposed by Vance et al.(1987) and the basal respiration according to the respirometrytitration method of Alef and NanniPieri (1995).The metabolic quotient (qCO2) and microbial quotient were determined according to the relation proposed by Anderson and Domsch (1990).
Data on air temperature, rainfall, air humidity (Figure 2A) and soil temperature (Figure 2B) on the days of evaluation are presented below and show variations between the periods analyzed (winter Data analysis was performed by descriptive statistics, in which means, standard deviation, maximum and minimum values and coefficient of variation were calculated.To compare the means, the Student's t-test was used at 5% probability.The analysis of variance (repeated measurements over time) and linear regression were used for analysis of the temporal variability with graphical representation made in the software SigmaPlot, version 11.0.

RESULTS AND DISCUSSION
The average FCO 2 flux in the evaluated periods (winter and summer) was significantly higher in the GS-10 area, with 2.37 µmol m -2 s -1 , compared to the other areas, with 1.69 µmol m -2 s -1 for BS and 1.10 µmol m -2 s -1 for GS-5 (Table 2).
We believe that the higher FCO 2 on the GS-10 area is related to the greater amount of plant residue on the soil, which is a result of the mechanized harvesting.The presence of mulch on the soil provides a habitat for microorganisms (Franchini et al., 2007) and serves as substrate for the microbial activity during the process of decomposition of the organic matter, thus reflecting in increased CO 2 emissions (Evanylo and Mcguinn, 2009).In addition, the minimum soil tillage provides favorable conditions for the development of microorganisms in the surface layer of the soil, which increases the microbial biomass and the FCO 2 (Matias et al., 2009).
The evaluation periods influenced the FCO 2 in the green sugarcane areas, being 51 and 18% higher in summer than in winter for the GS-5 and GS-10 areas, respectively, while for the burned sugarcane area, the variation of FCO 2 between periods was less expressive (Figure 3A and B).Studies have shown higher FCO 2 on the wettest period of the year (Xu and Qi, 2001;Kosugi et al., 2007;Song et al., 2013), which may be related to the greater microbial activity stimulated by the soil moisture and/or by the activity of the roots that are in the period of growth and development.Siqueira et al. (2011) verified that the maximum CO 2 emission occurred in the Brazilian Cerrado area in the rainy season (October-March), which was 14 times higher than the minimum emission obtained in the dry season (April-September).
The basal respiration (BR), in winter, was higher in the burned sugarcane with 109.67 µg CO 2 g -1 day -1 and, in summer, it was higher in the GS-5 area with 96.05 µg CO 2 g -1 day -1 , while in the GS-10 area it was lower both in winter and in summer with 50.94 and 40.67 µg CO 2 g -1 day -1 , respectively (Table 2 and Figure 3C and D).According to Lopes et al. (2013), balanced ecosystems tend to have lower rates of BR, as the vegetation cover on the soil, which is characteristic of these areas, provides material at different levels of decomposition and complexity of plant residues, thus resulting in lower levels of soil respiration.
The analysis of BR, in which the CO 2 emission from the soil was calculated by incubation process in laboratory, presented a divergent trend from the FCO 2 obtained in field (Table 2).It is worth mentioning that the FCO 2 calculates the resulting gas flux resulting from the soil microbial activity and plant roots, while BR takes into account only the respiration of microorganisms, which is one of the reasons for the difference of patterns between FCO 2 and BR.In addition, in the literature, the isolated evaluation of the BR can lead to misunderstandings, since it can be interpreted both in the beneficial sense to the soil, with an accelerated organic matter decomposition process reflecting on nutrient availability to plants, and in the unfavorable sense, with large CO 2 emission, which shows more losses than gains in carbon in the soil (Alves et al., 2011;Evangelista et al., 2013;Lopes et al., 2013).
The carbon of the microbial biomass (CMB) in winter showed no significant differences when compared between the three areas of sugarcane management, with the values of 184, 186 and 197 µg C g -1 day -1 for BS, GS-  ); Sm = Soil moisture (%).Means followed by the same lowercase letter in the column (management systems) and uppercase letter in the row (evaluation periods) do not differ by the Student's t-test at 5% probability.
5 and GS-10, respectively (Table 2 and Figure 4).However, in summer, the CMB was 6.5 and 5.6% higher in the GS-10 area when compared with the BS and GS-5 areas, respectively.Similar trend was found in the study of Mendonza et al. (2000), who detected values of CMB of 152.1 and 195.6 µg C g -1 day -1 in the 00-0.05m soil layer for the areas of burned sugarcane and green sugarcane, respectively, and Galdos et al. (2009) verified that the CMB was 2.5 times higher in the area of green sugarcane compared to the burned sugarcane.
The values of CMB corroborate with the values of FCO 2 evaluated in field, in which the GS-10 area showed the highest amount of microbial biomass, which reflects into a greater soil CO 2 flux (Table 2).Thus, a direct relation between these two attributes can be observed as the CO 2 emitted by the soil is essentially produced by the decomposition of organic matter from the action of microorganisms and the respiration of plant roots (Panosso et al., 2008).The study of Xu and Qi (2001) in soil with pine plantation proved the direct relationship between FCO 2 and CMB during the monitoring of both attributes in the period from J une to October, 1998.
The evaluation of the CMB between the periods of evaluation was significant, being greater in summer than winter in the three sugarcane areas evaluated (Table 2 and Figure 4), as the wet period is characterized by conditions of temperature and soil moisture that are more favorable to the soil microbial activity (Mendonza et al., 2000;Zornoza et al., 2007).The optimum humidity is approximately 60 to 80% and the optimum temperature is 30°C.However, the activity decreases as the soil temperature exceeds the optimum temperature (Evanylo and Mcguinn, 2009).
Soil moisture data ranged from 10 to 11% in winter and from 21 to 38% in summer (Table 2), and the soil temperature ranged from 20 to 22°C in winter and from 18 to 19°C in summer (Figure 2B), being the attributes suboptimal, which boosts the mulch factor in the stimulation of the soil microbial activity, mainly in the GS-10 area, which showed the highest CMB.A similar result was obtained by Mendonza et al. (2000), who verified higher CMB in the green sugarcane system than in the burned sugarcane, especially during the rainy season in study under yellow Argisol, which, according to the authors, together with the favorable climatic factors in the green sugarcane system, would have allowed the microbial biomass to find plenty of substrate for its development in the remaining mulch from the previous year.
The values of qCO 2 ranged from 0.10 to 0.63 µg CO 2 µg C-SMB day -1 (Table 2).The low value of qCO 2 indicated economy on energy usage, which reflects a more stable environment or closer to its equilibrium (Anderson and Domsch, 1990).The results show that the cultivation system significantly affected the qCO 2 of the BS and GS-10 areas, with lower qCO 2 in GS-10 (Figure 4), which, according to Partelli et al. (2012), is the evidence of the lower consumption of oxidizable carbon for the maintenance of microorganisms, thus configuring the equilibrium situation of the soil microbial activity.
In a study by Evangelista et al. (2013), the qCO 2 was 49.07%higher in the burned sugarcane area that in the green sugarcane, and, according to the authors, as the microbial biomass becomes more efficient in the use of the ecosystem resources, less CO 2 is lost by the respiration and a higher proportion of carbon is incorporated into the microbial tissues, which results in decreased qCO 2 .Kuwano et al. (2014), working with soils under different uses in Northern Paraná, observed that the sugarcane areas under the burning system presented the highest qCO 2 , probably as a result of the burning before harvest, which decreases soil carbon inputs; moreover, the disturbance of the soil through heavy tillage on the renovation of the sugarcane plantation disrupts the microbial community not only by the breakdown of soil aggregates, but also by the loss of water from the soil.
The qMIC, obtained by the relationship between CMB and OC, was similar in winter in the BS and GS-10 areas, with 6.93 and 7.35 µg CO 2 day -1 µg OC, respectively, which may be related to the similar levels of organic carbon present in these areas, with 2.94 and 2.59 g kg -1 for BS and GS-10, respectively (Table 2).In summer, the qMIC was higher in the GS-10 area, with 12.28 µg CO 2 day -1 µg OC, and lower in the burned sugarcane area, with 7.57 µg CO 2 day -1 µg OC (Figure 4), and a higher qMIC is an evidence of organic matter of better quality, more active and less recalcitrant, with greater availability of organic carbon to the soil microbial activity (Jenkinson, Ladd, 1981).
In some studies, FCO 2 was positively correlated with organic carbon (La Scala et al., 2000;Medeiros et al., 2011;Lenka, Lal, 2013).However, in this study, the GS-10 area showed lower organic carbon content and greater FCO 2 .It is possible that the high microbial activity in the GS-10 area has reduced the organic carbon content, as the increase of cycles of decomposition of soil organic matter by microorganisms will result in a low organic carbon content, which is more protected and stabilized within microaggregates (Lenka and Lal, 2013).Corroborating with the result of this study, Fang et al. (1998) detected higher CO 2 emission in regions with lower organic carbon content in soil under pine plantation.

Conclusion
1. Soil CO 2 flux was higher in the area of green sugarcane implemented for 10 years.
2. Soil CO 2 flux and microbial activity were higher in the wet period.3. The metabolic and microbial quotients showed a greater balance of the soil microbial activity in the area of green sugarcane implemented for 10 years.4. The presence of mulch on the soil stimulates the microbial activity and increases the soil CO 2 flux.

BFigure 2 .
Figure 2. Rainfall, temperature, air humidity (A) and soil temperature (B) on the days of evaluation of soil CO 2 flux in winter and summer.

Figure 3 .
Figure 3. Soil CO 2 flux -FCO 2 (A and B), basal respiration -BR (C and D) of the soil evaluated in winter and summer in the management systems of burned sugarcane (BS), green sugarcane implemented for 5 years (GS-5) and green sugarcane implemented for 10 years (GS-10) in Pradópolis, São Paulo, Brazil, 2011-2012.

Figure 4 .
Figure 4. Effect of different sugarcane management systems (burned sugarcane, green sugarcane implemented for 5 years and green sugarcane implemented for 10 years) in the metabolic quotient (qCO 2 ) and microbial quotient (qMIC) in the summer and winter periods in Pradópolis, São Paulo, Brazil, 2011-2012.

Table 2 .
Microbiological attributes and soil moisture evaluated in the winter and summer periods on the sugarcane management systems (burned sugarcane, green sugarcane implemented for 5 years and green sugarcane implemented for 10 years) in Pradópolis, SãoPaulo,  Brazil, 2011-2012.