C-LABELED GLUCOSE AND AMMONIUM IN SALINE ARABLE SOILS

Organic matter dynamics and nutrient availability in saline agricultural soils of the State of Guanajuato might provide information for remediation strategies. C labeled glucose with or without 200 mg kg of NH4 -N soil was added to two clayey agricultural soils with different electrolytic conductivity (EC), i.e. 0.94 dS m (low EC; LEC) and 6.72 dS m (high EC; HEC), to investigate the effect of N availability and salt content on organic material decomposition. Inorganic N dynamics and production of CO2 and 14CO2 were monitored. Approximately 60 % of the glucose-C added to LEC soil evolved as 14CO2, but only 20 % in HEC soil after the incubation period of 21 days. After one day, < 200 mg C was extractable from LEC soil, but > 500 mg C from HEC soil. No N mineralization occurred in the LEC and HEC soils and glucose addition reduced the concentrations of inorganic N in unamended soil and soil amended with NH4 -N. The NO2 and NO3 concentrations were on average higher in LEC than in HEC soil, with exception of NO2 in HEC amended with NH4 -N. It was concluded that increases in soil EC reduced mineralization of the easily decomposable C substrate and resulted in Ndepleted soil.


SUMMARY
Organic matter dynamics and nutrient availability in saline agricultural soils of the State of Guanajuato might provide information for remediation strategies. 14 C labeled glucose with or without 200 mg kg -1 of NH 4 + -N soil was added to two clayey agricultural soils with different electrolytic conductivity (EC), i.e. 0.94 dS m -1 (low EC; LEC) and 6.72 dS m -1 (high EC; HEC), to investigate the effect of N availability and salt content on organic material decomposition. Inorganic N dynamics and production of CO 2 and 14 CO 2 were monitored. Approximately 60 % of the glucose-14 C added to LEC soil evolved as 14 CO 2 , but only 20 % in HEC soil after the incubation period of 21 days. After one day, < 200 mg 14 C was extractable from LEC soil, but > 500 mg 14 C from HEC soil. No N mineralization occurred in the LEC and HEC soils and glucose addition reduced the concentrations of inorganic N in unamended soil and soil amended with NH 4 + -N. The NO 2 and NO 3 concentrations were on average higher in LEC than in HEC soil, with exception of NO 2 in HEC amended with NH 4 + -N. It was concluded that increases in soil EC reduced mineralization of the easily decomposable C substrate and resulted in Ndepleted soil.
Index terms: dynamics of inorganic N, emission of 14 CO 2 and 12 CO 2 , saline soils.

INTRODUCTION
Nowadays, there is an environmental and economic need to understand the role and destiny of N in different ecosystems. Nitrogen cycling is mostly controlled by biological activity and at the same time biological processes are affected by climate and physicochemical soil characteristics. In some extreme environments, such as saline soils, the high electrolytic conductivity (EC) inhibits microbial activity and organic matter decomposition and thus affects N cycling (Johnston & Guenzi, 1963;McCormick & Wolf, 1980;Bandyopadhyay & Bandyopadhyay, 1983;Zahran, 1997;Pathak & Rao, 1998). For instance, mineralization of maize and glucose were inhibited in alkaline saline soils with EC > 10 dS m -1 and large amounts of NH 4 + and NO 3 were immobilized within short periods of time, reducing N availability (Conde et al., 2005). Additionally, high concentrations of nitrite (NO 2 -) were accumulated in these soils when an easily decomposable substrate plus NO 3 were added (Vega-Jarquin et al., 2003). Pathak & Rao (1998) reported that ammonification and nitrification were inhibited by high salt concentrations and that, particularly the latter, was very sensitive to the presence of salts.
Plant growth and crop yields have decreased in some parts of the State of Guanajuato, Mexico, due to the excessive amounts of salts in the soil, and some parts have become uncultivable. The salt in soil increases since crops are irrigated with saline effluents and by inadequate soil practices. The Agriculture Department of the State of Guanajuato has started a project to investigate how increased salinity affects nutrient availability in the soil and how the addition of organic material might restore soil fertility. As part of this project, the effect of increased salinity on dynamics of inorganic N (NH 4 + , NO 2 -, NO 3 -) and organic material were investigated. Two agricultural soils with different electrolytic conductivity (EC), i.e. 0.94 dS m -1 (low EC; LEC) and 6.72 dS m -1 (high EC; HEC) were amended with or without 14 C-labeled glucose and with or without (NH 4 ) 2 SO 4 . The 14 Clabeled glucose is routinely used to determine effects of soil characteristics on mineralization of organic material (Saggar et al., 1999). The objective of this study was to investigate the effect of EC and inorganic N on the decomposition of organic matter in two soils of Guanajuato State, Mexico.

Sampling site
The sampling sites are located in Cuerámaro (HEC soil) and San Francisco del Rincón (LEC soil), State of Guanajuato, Mexico, at 1,726 and 1,804 m asl, with a mean annual temperature of 20.3 and 19.4 °C; and precipitation of 981 and 967 mm, respectively (Terrones-Rincón et al., 2000). Soils in the area originating from aluvio-coluvial deposits are clayey-loamy to clayey-sandy. The soil structure is granular in the top soil and organic matter contents range from 14 to 52 g kg -1 dry soil; EC in saturation extracts ranges from 0.9 to 6.7 dS m -1 . The cation exchange capacity (CEC) in soils ranges from 34.8 to 40.5 cmol c kg -1 dry soil while exchangeable Ca 2+ from 66 to 90 mg kg -1 , Mg 2+ from 17 to 26 mg kg -1 , Na + from 82 to 163 mg kg -1 , K + from 15.6 to 19.6 mg kg -1 , and extractable P from 23 to 32 mg kg -1 . CO 3 2in the exchangeable saturation extract ranges from 15 to 24 mg kg -1 , Clfrom 82 to 93 mg kg -1 and SO 4 2from 230 to 289 mg kg -1 (Castellanos- Ramos et al., 2000). Approximately 17 % of the study area is irrigated. The natural vegetation consists of species that can be used for cattle raising, such as Bouteloua chasei, Muhlenbergia spp., Prosopis spp., Camedrioteucrium chamaedrys, Bothriochloa spp., Buchloe dactyloides, Aristida divaricata, Liendilla lanuda, Bouteloua curtipendula, Acacia spp., Opuntia spp., and Uncaria tomentosa.
Two agricultural soils with different EC were sampled (Table 1). In each one, three plots of approximately 400 m 2 were defined from which 30 soil samples were collected by augering the top 0-15 cm layer. The 30 soil samples from each plot were pooled separately and characterized for a total of six soil samples, i.e., soil from two fields and three plots.

Aerobic incubation
The soil samples were passed through a 5 mm sieve and placed in drums containing a beaker with 100 mL distilled H 2 O to avoid desiccation. One contained 100 mL 1 mol L -1 NaOH solution to trap any CO 2 evolved for six days.
Sixty soil sub-samples (30 g) from each field and plot were filled in 120 mL glass flasks. Fifteen subsamples were amended with a solution containing 14 Clabeled glucose (approximately 1.48 MBq kg -1 ), 15 with 14 C-labeled glucose plus (NH 4 ) 2 SO 4 , 15 with (NH 4 ) 2 SO 4 and the others were treated with an equal amount of distilled H 2 O. The amount of water added resulted in a soil moisture content of approximately 50 % WHC, and the amounts of C and N added as glucose and NH 4 + were approximately 1,000 mg kg -1 C and 200 mg kg -1 N. Three flasks were chosen at random from each treatment. Soil was extracted (30 g) for inorganic N and 14 C with 120 mL 0.5 mol L -1 K 2 SO 4 solution to provide zero time samples. The samples were shaken for 30 min, filtered through Whatman No 42 paper® and stored until analysis at -20 °C (Jenkinson & Powlson, 1976).
The glass flasks were placed in 945 mL glass jars with 10 mL distilled H 2 O. One contained a vessel with 20 mL 1 mol L -1 NaOH solution to trap CO 2 evolved and another a vessel with 20 mL 2 % H 3 BO 3 solution to trap volatilized NH 3 . The jars were sealed airtight and stored in the dark for 21 days at 22 ± 1 °C. After 1, 3, 7 and 21 days, three jars selected at random from each treatment were opened, and the vessels containing NaOH and H 3 BO 3 were removed. An aliquot of 0.2 mL 1 mol L -1 NaOH solution was taken and analyzed for 14 C activity. The remaining NaOH and H 3 BO 3 solution was titrated with appropriate concentrations of H 2 SO 4 to determine CO 2 and NH 3 trapped, respectively. The soil was removed from the three flasks and 30 g was extracted with 120 mL 0.5 mol L -1 K 2 SO 4 solution. The samples were shaken for 30 min and filtered through Whatman N° 42 paper® and the inorganic and organic N and 14 C were measured as described for zero time samples (Conde et al., 2005).

Chemical analysis
The pH was measured in 1:2.5 soil/H 2 O suspension using a glass electrode (Thomas, 1996). Total C was determined by oxidation with potassium dichromate  (Kalembasa & Jenkinson, 1973), and inorganic C by adding 5 mL 1 mol L -1 HCl solution to 1 g air-dried soil and trapping CO 2 evolved in 20 mL 1 mol L -1 NaOH. Total N was measured by the Kjeldhal method using concentrated H 2 SO 4 , K 2 SO 4 and HgO to digest the sample (Bremner, 1996), soil particle size distribution by the hydrometer method as described by Gee & Bauder (1986) and CEC with the barium acetate method (Jackson et al., 1986).
The NH 4 + , NO 2 and NO 3 in the 0.5 mol L -1 K 2 SO 4 extracts were measured colorimetrically with an automatic Skalar San plus System (the Netherlands). The 14 C in the extracts was measured with a scintillation counter (Beckman LS6000SC, USA). The CO 2 in the 1 mol L -1 NaOH was determined by titration with 0.1 mol L -1 HCl (Jenkinson & Powlson, 1976). The WHC was measured in soil samples watersaturated in a funnel and left to stand overnight and defined by weight differences (Conde et al., 2005).

Statistical analysis
The cumulative production of 12 CO 2 and 14 CO 2 was regressed on elapsed time using a linear model forced to pass through the origin, but allowing different slopes (production rates) for each treatment. This approach was supported by the theoretical considerations that no 12 CO 2 and 14 CO 2 was produced at time zero and a control (flask without soil) accounted for atmospheric 12 CO 2 and 14 CO 2 .
The data of soil characteristics, concentration of NH 4 + , NO 2 and NO 3 -, and extractable 14 C were subjected to one-way analysis of variance using PROC GLM (SAS, 1989) to test for significant differences between treatments with Tukey's Studentized Range test. Significant differences between treatments for 12 CO 2 and 14 CO 2 production were determined using PROC MIXED (SAS, 1989).

RESULTS
Extractable 14 C and production of 14 CO 2 The CO 2 production rate was significantly higher in LEC soil amended with NH 4 + than in the unamended soil, but lower than in soil amended with 14 C labeled glucose or 14 C labeled glucose plus NH 4 + (p < 0.05) (Figure 1a, Table 2). The CO 2 production rate was significantly higher in HEC soil amended with 14 C labeled glucose or NH 4 + than in the unamended soil, but lower than in soil amended with 14 C labeled glucose plus NH 4 + (p < 0.05) (Figure 1b, Table 2). The CO 2 production rate was 1.3 times and significantly higher in LEC soil than in HEC soil (mean of all treatments) (p < 0.05).
The production of 14 CO 2 was greater in LEC soil amended with glucose than in soil amended with glucose plus NH 4 + (Figure 1c, Table 2). The production of 14 CO 2 was lower in HEC soil amended with glucose than in soil amended with glucose plus ammonium soil (Figure 1d, Table 2). The 14 CO 2 production rate was 1.4 times and significantly higher in LEC soil than in HEC soil (mean of all treatments) (p < 0.05).
The amount of extractable 14 C was similar in LEC soil amended with glucose and glucose plus NH 4 + over time (Figure 1e). A sharp drop was detected after the first day and a flattening out afterwards. Based on the concentration of 14 C in the soil extracts, approximately 150 mg C of the added glucose could not be detected one hour after its application in LEC soil and 100 mg glucose-C in HEC soil. Less than 10 mg kg -1 14 C soil was measured after 21 days. The extractable 14 C decreased more slowly in HEC soil than in LEC soil and the residual concentration of 14 C was > 68 mg kg -1 14 C in soil amended with glucose plus NH 4 + after 21 days (Figure 1f). The extractable 14 C decreased faster in HEC soil amended with glucose plus NH 4 + than in soil amended with glucose and the -N dry soil ( ), 1,000 mg kg -1 14 C-labeled glucose C dry soil ( ) or with 200 mg kg -1 NH 4 + -N dry soil plus 1,000 mg kg -1 14 C-labeled glucose C dry soil ( ) incubated aerobically at 22 ± 2 °C for 21 days. Bars are ± 1 STD and each point is the mean of n = 9. residual concentration was also higher in the latter, i.e., 245 mg kg -1 . 14 C. The amount of extractable 14 C was significantly higher in HEC than in LEC soil (mean of all treatments) (p < 0.05) ( Table 2). the maximum was lower, reached after seven days. The mean NO 2 concentration was similar in LEC and HEC soil (mean of all treatments) (p < 0.05) ( Table 2).

Inorganic N
The emissions of NH 3 from both soils < 4 mg kg -1 N soil and treatment had no significant effect on the mean NH 3 concentrations volatilized from both the LEC soil and HEC soils ( Table 2). The mean amount of NH 3 volatilized was similar for LEC soil and HEC soil (mean of all treatments) (p < 0.05).
The NH 4 + concentrations in the unamended LEC soil and HEC soil remained < 4 mg kg -1 NH 4 + -N and glucose addition decreased these values < 0.2 mg kg -1 NH 4 + -N (Figure 2a,b). The concentration of NH 4 + was lower in soil amended with glucose plus NH 4 + than in soil amended with NH 4 + alone ( Table 2). The mean concentrations of NH 4 + were similar in both soils (mean of all treatments) (p < 0.05).
The pattern of NO 2 concentrations in the unamended LEC soil and soil enriched with NH 4 + was similar, i.e., a small decrease over time, with concentrations < 2 mg kg -1 N (Figure 2c). The NO 2 concentration in the glucose-amended soil also decreased over time, though the values were higher than in the unamended soil and soil enriched with NH 4 + . In the glucose-amended soil plus NH 4 + , the NO 2 concentrations increased sharply to > 7 mg kg -1 NO 2 -N on day 1 and 3 and then decreased sharply to < 2 mg kg -1 NO 2 -N. The NO 2 concentrations in the unamended HEC soil and soil enriched with glucose showed a similar pattern, i.e. a small decrease over time, with concentrations < 2 mg kg -1 N (Figure 2d). The NO 2 concentrations in the glucose-amended plus NH 4 + increased to a maximum of 5 mg NO 2 -N after three days, and then decreased. In the NH 4 + -amended soil, the concentrations of NO 2 showed a similar pattern as in glucose plus NH 4 + amended soil, but The NO 3 concentration did not change significantly over time in unamended LEC, but decreased in the glucose-amended soil (p < 0.05) Table 2. The 12 CO 2 and 14 CO 2 production rate, the mean extractable 14 C (mg kg -1 C soil), mean amount of NH 3 volatilized and the mean concentration of NH 4 + , NO 2 and NO 3 in LEC and HEC soil ( Figure 2e). In the NH 4 + plus glucose and NH 4 + amended soil, NO 3 increased over time with the highest increase found in the NH 4 + amended soil. The NO 3 concentration did not change significantly over time in the unamended HEC soil, the glucose and glucose plus NH 4 + -amended soil, but increased in the NH 4 + -amended soil (p < 0.05) (Figure 2f). The mean concentration of NO 3 was lower in HEC soil than in LEC soil (mean of all treatments) (p < 0.05) ( Table 2).

Concentration of extractable 14 C
After one day, < 200 mg 14 C was extractable from LEC, but > 500 mg 14 C from HEC soil. After one day, 20 % of the added glucose was mineralized to 14 CO 2 , with an efficiency of 60 % for C (Payne, 1970), so only 53 % of the glucose was mineralized although 80 % could not be detected in the soil. Part of the 14 C-labeled glucose was taken up by the microorganisms in LEC soil without being metabolized. A similar lag was observed in HEC soil. Approximately 10 % of the added glucose was mineralized to 14 CO 2 , at an efficiency of 60 % for C, so only 27 % of the glucose was mineralized although 50 % could not be accounted for in HEC soil. The lag between glucose uptake and utilization for biosynthesis has often been reported (Payne & Wiebe, 1978;Bremer & van Kessel, 1990;Bremer & Kuikman, 1994). It indicates, as stated by Coody et al. (1986), that CO 2 evolution is a poor indicator of glucose uptake rates by soil microbes.
The NH 4 + addition reduced the amount of extractable 14 C and thus stimulated decomposition of 14 C-labeled glucose. Saline soils are often N depleted and the addition of NO 3 or NH 4 + to glucose-amended soil increases CO 2 production and a priming effect (as explained below) may also be observed (Conde et al., 2005). 12 CO 2 and 14 CO 2 production Approximately 9.4 % of the soil organic matter C was mineralized within 21 days in LEC and 7.7 % in HEC soil. The percentage of organic C mineralized was lowest in the HEC soil. This suggests that salts decreased microbial activity. The decrease in CO 2 production, however, was lower than reported by Pathak & Rao (1998), who found that when EC increased from 1.1 dS m -1 to 96.7 dS m -1 , the CO 2 respiration decreased from 2.1 to 0.9 g kg -1 CO 2 -C in Sesbania-amended soil after 90 days. However, they added salts to soil (NaCl and CaCl 2 ) to increase EC whereas the high EC in the HEC soil was the result of a slow salt increase due to irrigation. The microorganisms in the soils used by Pathak & Rao (1998) were not adapted to high salt contents and this might have strongly inhibited their activity. Salt addition to non-saline soils would require the adaptation of microorganisms to osmostic and/or specific ion stress and would therefore inhibit their activity more strongly.
Approximately 60 % of the glucose-14 C added to LEC soil was evolved as 14 CO 2 and 50 % in LEC soil amended with glucose plus NH 4 + . The percentage of mineralized glucose-C in the LEC soil was similar to values reported in literature. Bremer & Kuikman (1994) found that approximately 60 % glucose was mineralized in a sandy loam soil and silt loam soil after 35 days when ≥ 576 mg kg -1 of glucose C soil had been added.
The amount of glucose-14 C mineralized was only 37 % in HEC soil. The higher EC in HEC than in LEC soil presumably explained the lower amount of glucose 14 C respired to 14 CO 2 , although the effect of other soil characteristics on glucose decomposition can not be ruled out. Clay content has often been found to affect the decomposition of organic material fertilization (Sørensen, 1981;van Veen et al., 1985;Amato & Ladd, 1992) as does the pH (Saggar et al., 1999), but pH and clay were similar in both soils. The specific surface area of the clay, and the nature rather than the amount of the clay mineral has also been found to affect the decomposition of organic material (Saggar et al., 1996;Torn et al., 1997), but this effect is presumably more important in the long term than in short-term laboratory incubation experiments. Considering the above, it appears that EC was the factor that explained most of the reduction in mineralization of glucose added.
Compared to the unamended soil, the application of glucose increased CO 2 production to 784 mg kg -1 C in LEC soil and to 525 mg kg -1 C in HEC soil after 21 days, but the amount of 14 CO 2 produced was only 660 and 374 mg kg -1 C in LEC soil and HEC soil, respectively. The accelerated decomposition of unlabeled soil organic matter following the addition of organic material has often been referred to as a "priming effect", and has been a matter of controversy for many years (Brookes et al., 1990). However, the conditions for an apparent priming effect, such as nonuniformly labeled substrate or a great substrate addition, are absent in this experiment and the bicarbonate in the soil solution could not explain the differences either, as they were similar in both soils, i.e., 0.26 cmol c kg -1 HCO 3 -. Brookes et al. (1990) described situations where a true priming effect is mainly caused by an increased turnover of microbial cell C (after glucose addition) or by an increased decomposition of native soil organic matter (after addition of rye-grass). In the first situation, the new microbial biomass partly replaces the native, resulting in a greater production of unlabeled CO 2 . In the second, the new microbial biomass is added to the already present native biomass and the increase in CO 2 production results from an increased contact between the native soil organic matter and enzymes produced by a more active microbial population.
R. Bras. Ci. Solo, 33:857-865, 2009 Compared to the soil amended with NH 4 + , the application of glucose plus NH 4 + increased the CO 2 production to 364 mg kg -1 C in LEC soil and to 369 mg kg -1 C in HEC soil after 21 days, but the amount of 14 CO 2 produced was 515 and 520 mg kg -1 C for LEC soil and HEC soil, respectively. As such, the addition of NH 4 + to the glucose-amended soil induced a negative priming effect. The addition of different substances to the soils might cause not only an acceleration of mineralization or positive priming effect, but also a reduction or a negative priming effect. A negative priming effect has often been reported for N, e.g. N immobilization and a temporal N unavailability, but less often for C. It is difficult to indicate which mechanisms might contribute to a negative priming effect, but Kuzyakov et al. (2000) stated that when a negative priming effect occurs the microbial biomass switches from metabolizing soil organic matter to easily available C sources, stimulated by NH 4 + . Another possibility is that the addition of an easily decomposable substrate plus NH 4 + induced the growth of an active glucose-decomposing population that inhibits the activity of a passive population decomposing soil organic matter. It might also be that the addition of NH 4 + itself inhibited microbial activity by the formation of NH 3 . NH 3 is toxic and known to inhibit microbial activity.

Inorganic N
The factors that normally most affect NH 3 volatilization are concentrations of NH 4 + , pH and water content (Kirchmann & Witter, 1989). Both soils had similar water contents, pH and mean NH 4 + concentrations so the mean amounts of NH 3 volatilized were similar (Table 2).
It has often been reported that nitrification and especially oxidation of NO 2 to NO 3 is inhibited by high salt concentrations (Darrah et al., 1987;Low et al., 1997;Pathak & Rao, 1998). However, in the experiment reported here the mean NH 4 + and NO 2 concentrations were similar so it appeared that the nitrification process was not inhibited.
The concentration of inorganic N (sum of NH 4 + , NO 2 and NO 3 -) decreased over time in HEC soil amended with NH 4 + , but not in LEC soil (Figure 3). The amount of N lost through NH 3 volatilization was similar in both soils so HEC was N-depleted, but not the LEC soil. Additionally, the concentration of inorganic N in HEC amended with glucose plus NH 4 + was lower than in LEC soil, i.e. N immobilization was higher in HEC than in LEC soil.

CONCLUSIONS
1. Organic matter decomposition and N mineralization were most affected in the HEC soil.
2. In HEC soil, the addition of NH 4 + led to N immobilization.
3. NH 4 + and NO 2 oxidation was not inhibited in the HEC as it was in the LEC soil. grant-aided support from CONACyT and CONCYTEG.