HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (16) Citing Articles via Google Scholar Google Scholar Articles by Sierra, J. Articles by Marbán, L. Search for Related Content PubMed Articles by Sierra, J. Articles by Marbán, L. GeoRef GeoRef Citation Agricola Articles by Sierra, J. Articles by Marbán, L.

DIVISION S-3-SOIL BIOLOGY & BIOCHEMISTRY

Nitrogen Mineralization Pattern of an Oxisol of Guadeloupe, French West Indies

^a Unité Agropédoclimatique, INRA Antilles-Guyane, Domaine Duclos, Prise d'Eau, 97170 Petit Bourg, Guadeloupe (FWI), France
^b LAQUIGE, CONICET, Velasco 847, 1414 Buenos Aires, Argentina

sierra{at}antilles.inra.fr

	ABSTRACT

TOP ABSTRACT INTRODUCTION Materials and methods Kinetics and Temperature... Results and discussion Conclusions REFERENCES

 
It is generally assumed that Oxisols have a low level of soil organic matter (SOM) and a high rate of SOM turnover due to the high temperatures in the tropics. This work was carried out to test this assumption by analyzing the kinetics and the temperature–moisture response of N mineralization and nitrification in a neutral (pH 6.2, organic N 1.5 g kg^-1, bare for 10 yr) and an acid (pH 4.7, organic N 2.3 g kg^-1, bare for 1 yr) Oxisol. Samples of these soils were incubated using a factorial design of temperature (20, 30, 40, and 50°C) by soil water (30, 200, and 1500 kPa). The mineralization pattern was related to measurements of microbial biomass, SOM light-fraction, and the number of nitrifiers. Mineralization increased continuously with the increase of temperature and water content. No nitrification was observed at 50°C and at 1500 kPa. Although mineralization was well described by the double exponential model, nitrification showed a linear pattern with time. The size of the mineralizable N pool and N mineralization were different between the two soils, but microbial activity and microbial biomass were almost identical in both soils. Mineralization was very slow at 20 to 30°C (constant rate 10^-4–10^-5 d^-1) and was limited by a relatively small (light-fraction C = 8–10% of organic C) and low active mineralizable fraction. At >40°C, the constant rate of mineralization (10^-3 d^-1) was equivalent to those found in temperate soils at 25 to 35°C. Our findings do not support the assumption of a fast SOM turnover at high temperatures. Nitrifiers constitute a small population (20 cells g^-1) well adapted to changes in moisture and substrate availability. However, NH⁺₄ was always present in soils, which was associated with the low number of nitrifiers. Microbial biomass decreased sharply with an increase in temperature and was not related to N mineralization. Thermal denaturation or changes in microbial population could cause this decrease.

Abbreviations: AS, acid soil • NS, neutral soil • SOM, soil organic matter

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Materials and methods Kinetics and Temperature... Results and discussion Conclusions REFERENCES

 
OXISOLS

constitute 830 million ha in the tropical regions, which represents 25% of the total tropical land (Sanchez and Logan, 1992). It is well known that changes in these soils from the native tropical rainforest or semideciduous forest to arable crops induce rapid depletion of soil fertility (Albrecht et al., 1992; Lal and Ragland, 1993). Although SOM and N diminish under continuous cultivation from their initial levels under forest conditions, they may remain relatively high even after several years of agriculture and similar to those observed in temperate soils of comparable land use (Greenland et al., 1992). However, as pointed out by Sanchez and Logan (1992)(p. 44), "many Oxisols with similar SOM contents (to those found in Mollisols of temperate regions) will definitely not be able to keep one corn plant alive without fertilizer additions." Although Al toxicity or acidity may eventually affect crop production in Oxisols, "soil fertility depletion" appears to be associated with a decrease of the most readily mineralizable SOM rather than the total SOM or the total pool of nutrients (Chotte et al., 1994). This is particularly the case of Oxisols having an udic moisture regime, as in the Caribbean region, where erosion and compaction are not major soil constraints (Cabidoche, 1996).

There is now sufficient evidence that a reasonable balance of organic and inorganic (fertilizer, lime) inputs is necessary to maintain or increase crop production and to develop sustainable options for agriculture in Oxisols (Sanchez et al., 1989; Lal and Ragland, 1993). For this, a better knowledge of SOM dynamics is needed to estimate the amount of N that mineralizes under tropical conditions and to calculate the correct N input for crops. Some models have been used to evaluate the turnover of SOM in tropics on the long-term scale (Greenland et al., 1992), but there is little information concerning the estimate of N availability at the short-term scale of a crop season. Most of the available information on soil N mineralization has been obtained in temperate regions, in soils usually well supplied with bases and with microbial populations dominated by bacteria. In the case of the tropics, especially in Oxisols, several questions remain unanswered: (i) Is the mineralization–immobilization pattern different in acid Oxisols with microbial populations dominated by fungi having a higher C/N ratio? (Sanchez et al., 1989); (ii) How do high temperatures of the tropics and soil acidity affect the mineralization–nitrification pattern (or, is the soil ratio modified by soil conditions)?; (iii) Are the models currently used in temperate soils able to describe N mineralization in Oxisols? This work was carried out to answer these questions. Our goal was to analyze the kinetics and temperature–moisture relationships of N mineralization and nitrification in acid and neutral-pH Oxisols, and to relate them to some soil biological parameters (microbial biomass C and nitrifiers) and to the light-fraction of SOM.

	Materials and methods

TOP ABSTRACT INTRODUCTION Materials and methods Kinetics and Temperature... Results and discussion Conclusions REFERENCES

 
Soil Characteristics and Laboratory Incubations
The soils used in this study are classified as Typic Eutroperox and are located at the Duclos Experimental Station of the Institut National de la Recherche Agronomique in Guadeloupe, French West Indies (16° 15' N, 61° 40' W). At this site the mean annual air temperature is 25.5°C (26.7°C for the hottest month and 23.4°C for the coldest month), and the mean annual rainfall is 4000 mm. The soils were selected based on soil pH (1:2.5 soil/water) within the first 0.1 m: 6.2 (neutral soil [NS], limed in 1993) and 4.7 (acid soil [AS]). The NS was located in a 200-m² plot that was bare for 10 yr before this study; AS was in a 40-m² plot under tropical grass [Dichanthium aristatum (Poir.) C.E.] but was bare for 1 yr before this study. Other characteristics of the 0.1-m surface layer are: clay fraction (<2 mm), 78% (dominated by kaolinite and halloysite); organic C (Elemental analyzer, Thermoquest Italia SPA, Rodano, Italy), 14.3 g kg^-1 for NS and 22.9 g kg^-1 for AS; total N (Elemental analyzer, Thermoquest Italia SPA), 1.5 g kg^-1 for NS and 2.3 g kg^-1 for AS; cation-exchange capacity, 14 cmol kg^-1 (55% of Ca²⁺) for NS and 8 cmol kg^-1 (40% of Ca²⁺) for AS. Soil samples were taken from the upper 0.1-m layer on a 1-m² subplot at the center of each plot. The samples were gently dried at room conditions (30°C, relative air humidity >80%) for 3 d to attain a gravimetric water content corresponding with 1800 kPa. The soil samples were not sieved (aggregate size <0.01) to avoid artifacts associated with soil disturbance (Sierra, 1996).

A first experiment of laboratory incubations consisted of a factorial design of temperature (20, 30, 40, and 50°C) by soil water (30, 200, and 1500 kPa) for each soil. The experiment was run in duplicate. The selected levels of soil moisture and temperature are within the range of values observed in our soil. For a bare soil located at the same Experimental Station, Bussiere and Cellier (1993) reported that at 0.02 m, the mean temperature was 31°C and the mean maxima and the mean minima temperatures were 35.5 and 25.6°C, respectively; they also reported that extreme values of maxima daily temperature such as 45 to 47°C were frequent in their soil. For the same soil under grass (standard conditions of meteorological station) the mean temperature at 0.1 m is 27.4°C and the mean maxima and the mean minima temperatures are 29.8 and 25.3°C, respectively (INRA, 1996). These values are similar to those reported by Myers (1975) for a tropical soil of Australia. This author observed that the maxima daily temperature at 0.025 m ranged from 41 to 52°C.

The incubation technique was that of Stockdale and Rees (1995), which is based on the principles of aerobic incubations described by Hart et al. (1994). Soil samples of 500 g (oven-dry basis) were placed in plastic buckets (0.16-m diameter, 0.12 m high) and distilled water was added to obtain the required soil moisture of each treatment by knowing the retention curve of each soil. Buckets were opened every 2 d, aerated for 5 min, and water content checked by weighing and adjusted as needed (water loss was always <2% of the initial moisture). Two subsamples (20 g dry soil) from each bucket were taken for inorganic N analysis after 0, 4, 10, 18, 25, 32, and 57 d of incubation. Nitrate-N and NH⁺₄–N were extracted with 100 mL of 0.5 M KCl by 2-h shaking. Preliminary work showed that 2-h shaking was enough to extract the entire mineral N from our clayey soils. The analyses were performed colorimetrically with a Technicon autoanalyzer (Technicon Industrial Systems, Tarrytown, NY) using the hydrazine reduction method for NO^-₃–N (Kampshake et al., 1967) and the phenol-nitroprusiate method for NH⁺₄–N (Kaplan, 1965). A second experiment with N-amended soils was run at the same time as Exp. 1. For this, 40 mg NH⁺₄–N kg^-1 as (NH₄)₂SO₄ was added to two 500-g samples of each soil and thereafter incubated at 30°C and 30 kPa. Ammonium was applied with distilled water to obtain the required soil moisture and NH⁺₄ concentration. Soil incubation and the measurements of NO^-₃–N and NH⁺₄–N were made as described above.

This incubation technique has three advantages in relation to the aim of this study compared with the conventionally used leaching methods:

1. Leaching may affect the mineralization–immobilization pattern in Oxisols. By comparing the leaching method proposed by Stanford and Smith (1972) and a no leaching method similar to that used in the present study, Mary and Recous (1994) observed that the net N mineralization was greater for the leaching method than for the no leaching method. They concluded that loss of C compounds by leaching reduced N immobilization. This may induce artifacts in soils having a low level of available C for microorganisms such as in Oxisols (Chotte et al., 1994).
   
2. The leaching method is difficult to use in experiments with soil samples amended with a soluble salt as in our Exp. 2.
   
3. The leaching method is also difficult to use with undisturbed soil because a large volume of solution (>1000 mL; e.g., Cabrera and Kissel, 1988a) has to be used to remove all of the inorganic N from soil aggregates. This is particularly important in Oxisols because a great amount of the inorganic N is present as NH⁺₄–N (Mench and Clairon, 1991; this study), which may be adsorbed by the clay particles within aggregates.
   

The light-C fraction of SOM, the microbial biomass C, and the amount of nitrifiers were determined prior to the experiments (before drying for microbial biomass and nitrifiers) and at the end of the first experiment for all the soil–temperature–moisture treatments. Nitrifiers were also determined at the end of the second experiment. The analyses of light-C fraction of SOM and microbial biomass C were made in duplicate; the number of nitrifiers was measured in triplicate. Two density fractions of SOM were isolated (Alvarez et al., 1998): the first was obtained by flotation in a BrH₄–ethanol solution (density 2 Mg m^-3), the second was obtained by flotation in CCl₄ (density 1.6 Mg m^-3). Organic C was determined by wet digestion. The biomass C was determined by the fumigation–extraction method as described by Vance et al. (1987). Briefly, moist samples of soil of each treatment were fumigated with ethanol-free CHCl₃ for 16 h at 20°C under vacuum. After fumigant removal, the fumigated samples were extracted by 2-h shaking with 0.5 M K₂SO₄, and the organic C of the soil extracts was measured using an Elemental Analyzer (Thermoquest Italia SPA). Control nonfumigated samples were extracted at the time of fumigation of the other samples. Biomass C was calculated by subtracting the extractable C in the nonfumigated sample from the extractable C in the fumigated sample. No factor was used to correct these values. The number of nitrifiers was measured by the most probable number method (Schmidt and Belser, 1994) with three replicates by dilution level and incubating them at 25°C in the dark. As recommended by these authors, and taking into account the low number of nitrifiers in our soils, the observations were made during 6 wk with continuous monitoring to attain maximum counts.

	Kinetics and Temperature–Moisture Response Models

TOP ABSTRACT INTRODUCTION Materials and methods Kinetics and Temperature... Results and discussion Conclusions REFERENCES

 
The fit of N mineralization and nitrification data was performed in two steps using the nonlinear iterative procedure of Bard (1974). In the first step we analyzed the kinetics and temperature–moisture response of the processes and tested the validity of some simplifications to describe N mineralization. We fitted all of the results of each soil (time, temperature, and moisture response) in the second step.

First Step
Experimental data of N mineralization was fitted using the double exponential model (Deans et al., 1986); the equation is:

(1)

where N_min (mg N kg^-1) is the cumulative N mineralized after time t; N₁ and N₂ (mg N kg^-1) are the recalcitrant and the labile organic N fractions decomposing at specific rates k₁ and k₂ (d^-1). The fit was realized under the imposed condition that N₁ and N₂ do not vary among treatments. This condition has a biophysical basis because if N₁ and N₂ are organic pools present in soil at the beginning of the incubation; then they depend on the soil but are not affected by treatment conditions.

For each soil, the k₁ and k₂ estimated using Eq. [1] were fitted with the Q10 function:

(2)

where k (d^-1) is k₁ or k₂, A (d^-1) is an empirical coefficient, T (°C) is the temperature, and Q10 is a temperature coefficient corresponding with the ratio of constant rates at temperatures differing by 10°C. To verify whether Q10 depended on soil moisture, the fit was performed individually for each moisture treatment and for the pooled data. For both soils and both constant rates, Q10 did not differ significantly between moisture treatments (P < 0.05), so that they may be assumed as being independent of soil moisture by taking a single value for all moisture treatments.

In order to assess the effect of soil moisture on N mineralization, we considered the coefficient A of Eq. [2] as a function of soil moisture (Sierra, 1997). The functions were: for k₁

(3)

and for k₂

(4)

where B (d^-1), C (d^-1 kPa^-1), D (d^-1 kPa^-1), and E are empirical constants, and (kPa) is the water potential. Note that the response to moisture was different for k₁ (linear) and for k₂ (power function).

Nitrification showed a linear relationship with time

(5)

where N_nit (mg N kg^-1) is the cumulative N nitrified after time t, and k_nit is the rate of nitrification (mg N kg^-1 d^-1). The response of k_nit to temperature 40°C was also described by the Q10 function; nitrification was inhibited at 50°C. The equations for temperature response were

(6a)

(6b)

where F (d^-1) is an empirical constant. The Q10 of nitrification were not moisture dependent (P < 0.05). Note that the response between 40 and 50°C is not known.

The coefficient F was considered to depend on soil moisture so we propose the following equation to relate nitrification to water potential

(7)

where G (d^-1) and H (d^-1 kPa^-1) are empirical constants. For = 1500 kPa, F 0 and k_nit 0, which describes nitrification inhibition at the lowest soil moisture tested in this study.

Second Step
Based on the analysis performed in the first step, we fitted the results of N mineralization and nitrification of each soil as a whole. The same assumption described above concerning N₁ and N₂ was used. In addition, we considered Q10 as independent of soil moisture. At each time, NH⁺₄–N concentration was calculated as the difference between N_min (Eq. [1]) and N_nit (Eq. [5]).

	Results and discussion

TOP ABSTRACT INTRODUCTION Materials and methods Kinetics and Temperature... Results and discussion Conclusions REFERENCES

 
Kinetics of Nitrogen Mineralization: A Comparison with Temperate Soils
For the range of the factors tested in this study, N mineralization increased continuously with the increase of temperature and soil moisture (Fig. 1) . Although the same pattern of response to temperature and moisture was observed in both soils, AS mineralized 20 to 30% more than NS (data not shown). From the estimates of the parameters of the fitted model, the greater N mineralization observed in AS may be explained by a greater mineralizable N pool but not by a higher response to temperature and moisture (Table 1) . Although N₁ and N₂ were significantly different among soils, Q10 and the coefficients of the moisture functions were not different. The double exponential model described well the experimental data of N mineralization in both soils (Fig. 1; , for NS and AS). Some minor discrepancies between observed and simulated data were found for the lower rates of mineralization; in these cases the model underestimated N mineralization (Fig. 1). This was particularly noticeable at 20°C and at 1500 kPa; for these treatments N mineralization increased rapidly in the first period but became more stable than that predicted by the model. The different values of Q10 estimated for k₁ and k₂ using Eq. [2] (R² > 0.96, P < 0.05, for both rates and both soils) indicate that these constant rates diverge in their response to changes in temperature. This is consistent with the biological meaning of N₁ and N₂. In fact, as N₂ is a more active organic pool for all temperature and moisture conditions, its response to changes in these variables is less (Q10 is less) than that of the recalcitrant N₁ pool (Table 1). For this reason, at 30 kPa, k₁ increased 70 times and k₂ increased only 25 times between 20 and 50°C (Fig. 2) . Equations [3] and [4] described well the response of k₁ and k₂ to soil moisture (R² > 0.98, P < 0.05, for both rates and both soils). The response to soil moisture was lower than that observed for temperature but the behavior of k₁ and k₂ was similar. For example, at 50°C, k₁ increased four times and k₂ increased three times between 1500 and 30 kPa.

View larger version (35K): [in this window] [in a new window]   Fig. 1 Kinetics of N mineralization and nitrification for the neutral soil Oxisol incubated at different temperature and water potential. Only N mineralization is presented for 50°C and 1500 kPa because no nitrification was observed in these treatments. Vertical bars indicate standard error. See text for the coefficient of determination of each fit

View this table: [in this window] [in a new window]   Table 1 Fitted parameters for the kinetics and temperature–moisture response of N mineralization and nitrification.

View larger version (17K): [in this window] [in a new window]   Fig. 2 Temperature response of the constant rates of N mineralization (k1 and k2) and nitrification (knit) for the NS Oxisol. The observed data were fitted with the Q10 equation. Vertical bars indicate estimation error. See text for the coefficient of determination of each fit

As the estimated mineralizable N pools are specific to the fitted model, a comparison of our results with those reported for temperate soils is only feasible when the double exponential model was used. Some data are available in literature for agricultural soils. For four U.S. Mollisols, Cabrera and Kissel (1988a) reported a range of N₁ values from 160 (sandy loam) to 280 mg N kg^-1 (silty clay loam). For N₂ the values ranged from 13 (silty clay loam) to 36 mg N kg^-1 (silty clay loam). Nordmeyer and Richter (1985) reported values in German soils varying from 128 to 179 mg N kg^-1 for N₁, and from 13 to 24 mg N kg^-1 for N₂. Deans et al. (1986) used the double exponential model to analyze results presented previously. They reported values of N₁ ranging from 70 to 800 mg N kg^-1 and from 10 to 100 mg N kg^-1 for N₂. Although our estimates of N₁ and N₂ are of the same order of magnitude than those cited in the literature for temperate soils, it seems that our N₁ values (Table 1) have to be placed in the lower range of values. This is also supported by the low levels of the light-fraction of SOM found in this study (Table 2) . The fraction <2 Mg m^-3 currently accounts for 10 to 30% of the total organic C in agricultural soils of temperate regions (Sierra, 1996; Alvarez et al., 1998). The low value of the 10-yr bare NS is not surprising because the lack of C input in recent years. In contrast, the light-fraction of AS is relatively small considering that this soil was under grass except for the year prior to this study. The light-fraction of SOM was not affected by temperature–moisture treatments, and the initial values did not differ from the final values (Table 2). This disagrees with previous work in temperate soils that indicated a good relationship between the light-fraction of SOM and N mineralization in short-term incubations (Sierra, 1996; Alvarez et al., 1998). However, the light fraction was more sensitive than the total organic C to changes in land use. The ratio NS C/AS C was 0.47 for the light fraction (Table 2) and 0.62 for the total organic C. As discussed above, this reflects soil fertility depletion in Oxisols (Chotte et al., 1994). It is interesting to remark that this depletion of soil C was not followed by a decrease in microbial activity (e.g., k₁ and k₂) or, as presented below, in the size of microbial biomass.

View larger version (18K): [in this window] [in a new window]   Fig. 3 Comparison between the temperature response of the constant rate of N mineralization k1 estimated in the present study for the NS Oxisol (30 kPa) and in two temperate soils. Data from the literature were obtained at 30 kPa. Data of Cabrera and Kissel (1988b) were calculated from their constant rate k1 estimated with the double-exponential model and the Q10 value assumed by these authors based on the results of Stanford et al. (1973)

The lack of nitrification at 1500 kPa was surprising because soils having a strong daily moisture variation frequently show significant nitrification even at water potential lower than 1500 kPa (Sierra, 1996). In our soils, the water potential of the 0- to 0.1-m layer may change from 30 to 1500 kPa within a few hours after a rainfall because of the high evapotranspiration rate (J. Sierra, unpublished data, 1998). Nitrification did not present a lag phase even if prior to the beginning of the incubations the soils were at 1800 kPa (Fig. 1). This suggests a rapid response of nitrifiers to the change of soil moisture from 1800 kPa to 30 or 200 kPa and implies that lower water potential may adversely affect nitrifier activity but not kill all nitrifying bacteria. This result is useful to understand the flush of nitrification and NO^-₃ leaching following tropical rainfall events (Mench and Clairon, 1991).

The Q10 of nitrification estimated with Eq. [6a] (R² > 0.98, P < 0.05, for NS and AS) were almost identical in both soils (Table 1). This disagrees with the current results reported in the literature on an increase of nitrification with liming (i.e., Bramley and White, 1989), but agrees with the significant nitrification at pH 4.5 observed by others (Joergensen et al., 1990; Alves et al., 1993) and also with the finding of Hankinson and Schmidt (1988) who detected the presence of an acidophilic Nitrobacter in an acid forest soil.

Ammonium was always present in our soils (Fig. 1). As N mineralization was higher in AS but nitrification was similar in both soils, NH⁺₄ concentration was slightly higher in AS for all of the temperature–moisture treatments (data not shown). It may be observed in Fig. 1 that 95% of the cumulative N mineralized at 40°C and 30 or 200 kPa was nitrified at the end of the experiment; in contrast, only 70% was nitrified at 30°C and 30 or 200 kPa. This could suggest that for the 57-d period of the experiment, the rate of NH⁺₄ release was not the limiting factor for nitrification but was principally affected by temperature and soil moisture. However, this picture is not sufficiently clear to explain the change of the rate of nitrification when NH⁺₄ was added to the soil. In the second experiment, we observed that the rates of nitrification at 30°C and 30 kPa for the first 18 d (2.41 mg N kg^-1 d^-1 for NS and 2.49 mg N kg^-1 d^-1 for AS) were approximately 10 times higher than those found in the unamended soils (Exp. 1, 0.26 mg N kg^-1 d^-1 for NS and 0.24 mg N kg^-1 d^-1 for AS). After Day 18, the rate of nitrification of the amended soil was similar to that of the first experiment, with a minimum of 5 to 6 mg NH⁺₄–N kg^-1 always present in the soils up to the end of the incubation. After Day 25, the differences in cumulative inorganic N between the amended (Exp. 2) and the unamended soils (Exp. 1) were constant and 40 mg N kg^-1 (data not shown); therefore we assumed that the rate of NO^-₃–N immobilization was negligible in relation to NO^-₃–N release.

The presence of NH⁺₄ in Exp. 1 and at the end of Exp. 2 could be associated with the low number of nitrifiers found in our soils; the mean value for the treatments with nitrification was 17 cells g^-1 (range 13–20 for both soils). Similar results were obtained for the initial measurements and at the end of the second experiment. No nitrifiers were detected at the end of the 50°C and 1500 kPa treatments. These results agree with a previous study with AS (Prior and Beramis, 1990). They found a low nitrifier population in the unamended soil (<10² cells g^-1) and an increase of the population when the soil was amended with sewage sludge (10³–10⁴ cells g^-1). In our second experiment nitrifiers were measured only at the end of the incubations, so we do not know whether changes in the nitrification rates corresponded with changes in nitrifier population or in nitrifier activity. In spite of this, it is clear that nitrifiers in the studied Oxisols are well adapted to rapid changes in soil moisture. During the dry periods a very low number of nitrifiers remain in the soil in an inactive or dormant phase. When the soil is rewetted, they respond rapidly and a few bacteria suffice to develop an activity similar to that observed in soils of temperate regions. The dormant population also responds rapidly when water and NH⁺₄ are added together. Because of the low population density, only NH⁺₄ present surrounding colonies and moving by diffusion would be available for nitrification. This would explain why the rate of nitrification increased in the amended soils, but NH⁺₄ was not fully nitrified.

Microbial Biomass
The initial biomass C was 95 mg C kg^-1 for NS and 78 mg C kg^-1 for AS, which is in the range of values reported for similar soils in the humid tropics (i.e., Luizao et al., 1992; Grissi et al., 1998; the comparison was made considering the correction factor used by these authors). Biomass C decreased with temperature and increased with soil moisture (Fig. 4) , and for this reason no reliable relationship was found between biomass C and any rate constant estimated in this study (Fig. 5) . For example, for a biomass of 45 mg C kg^-1 the rate k₁ ranged from 4 x 10^-5 d^-1 (20°C and 1500 kPa) to 2 x 10^-3 d^-1 (40°C and 30 kPa). Although it is well known that soil drying induces a decrease in microbial biomass (i.e., Luizao et al., 1992), only few works report the effect of high temperature on soil microbial biomass (Joergensen et al., 1990; Grissi et al., 1998). Joergensen et al. (1990) monitored changes of microbial biomass in a soil incubated at 15, 25, and 35°C. They found that at the lower temperatures microbial biomass decreased slowly with time, but at 35°C biomass fell abruptly even if microbial respiration was enhanced. They concluded that temperature had two effects: increasing the release of substrate for respiration and enhancing microbial death by thermal denaturation. In addition, calculations by these authors showed that most of the respired C came from SOM. In our experiment, the behavior of biomass C was partially different than that found by Joergensen et al. (1990): most of the decrease in the final biomass C occurred at the lower temperatures (20–30°C), and the decrease was weak between 40 and 50°C (Fig. 4). Therefore, even if a thermal denaturation may not be discarded to explain our observations, it seems that other factors also were involved in the decrease of biomass C. A complementary hypothesis is that changes of temperature induce changes in microbial population differing in their sensitivity to CHCl₃, and then the decrease in the measured biomass C might be an artifact due to incomplete kill of the soil microorganisms for the higher temperature treatments. It is well known that fungi are more tolerant than bacteria to high temperature (Dommergues and Mangenot, 1970), and nitrifier bacteria showed a great susceptibility to the highest temperature tested in this study. Therefore, a change of microbial population with temperature might be associated with the change in the fungi/bacteria ratio. In spite of this, it is not clear how this hypothetical change could affect our results because most of the studies indicate that the efficiency with which CHCl₃ fumigation kills fungi is higher than for bacteria (i.e., Vance et al., 1987; Toyota et al., 1996), although the contrary was also obtained (Ingham and Horton, 1987).

View larger version (30K): [in this window] [in a new window]   Fig. 4 Microbial biomass C measured at the beginning and at the end of the Exp. 1 for the neutral soil (NS) and the acid soil (AS) Oxisols

View larger version (16K): [in this window] [in a new window]   Fig. 5 Relationship between the constant rate k1 estimated for the neutral soil (NS) Oxisol and the microbial biomass C at the end of Exp. 1

	Conclusions

TOP ABSTRACT INTRODUCTION Materials and methods Kinetics and Temperature... Results and discussion Conclusions REFERENCES

The inhibition of nitrification at high temperature and low water content implies that the abrupt sequences of soil drying–rewetting and the high temperatures in the surface layer, typical characteristics of udic Oxisols, may be the major factors controlling nitrification in the field. As N mineralization is less sensitive to these soil conditions, a great part of the mineral N in the surface layer had to be present as NH⁺₄–N. This finding may be helpful to adjust N input management in order to diminish NO^-₃ leaching in humid tropics.

Most of the differences between our results and data reported for temperate soils were quantitative. In this way, models of N mineralization proposed in the literature might be adapted in Oxisols by taking into account principally (i) the shift in the biological response to temperature and (ii) the description of nitrification as an individual process that is limited simultaneously by substrate, environmental conditions, and the number of nitrifiers and is inhibited at high temperature and low water potential. In contrast, soil pH had a minor effect, apparently because of microbial adaptation.

If our observation concerning the decrease of microbial biomass C with temperature resulted because of a change in the microbial population, the estimate of this organic pool by the fumigation–extraction method becomes unreliable and prevents its use in models of SOM turnover in tropics. Further work is necessary to know the dynamics of microorganisms in Oxisols and a first step is to quantify the biomass and the activity of soil fungi and bacteria as a function of environmental conditions and C and N input.

	ACKNOWLEDGMENTS

 
We gratefully acknowledge D. Gougougnan (INRA), M. Beramis (INRA), and V. Richter (LAQUIGE) for technical assistance, and S. Adiku (University of Ghana) for reviewing the English manuscript.

Received for publication October 4, 1999.

	REFERENCES

TOP ABSTRACT INTRODUCTION Materials and methods Kinetics and Temperature... Results and discussion Conclusions REFERENCES

 

• Addiscott T.M. Kinetics and temperature relationships of mineralization and nitrification in Rothamsted soils with differing histories. J. Soil Sci. 1983;34:343-353.
• Albrecht A., Brossard M., Chotte J.L., Feller C. Les stocks organiques des principaux sols cultivés de la Martinique (Petites Antilles). (In French, with English abstract.) Cah. Orstom 1992;27:23-36.
• Alvarez C.R., Alvarez R., Grigera M.S., Lavado R.S. Associations between organic matter fractions and the active soil microbial biomass. Soil Biol. Biochem. 1998;30:767-773.
• Alves B.J.R., Urquiaga S., Cadish G., Souto C.M., Boddey R.M. In situ estimation of soil nitrogen mineralization. In: Mulongloy K., Merckx R., eds. Proc. Int. Symp. Lab. of Soil Fertility and Soil Biology, Katholieke Univ. Leuven and IITA. Chichester, UK: Wiley, 1993:173-180 Leuven, Belgium. 4–6 Nov. 1991..
• Bard A. Non-linear parameter estimation. New York: Academic Press, 1974.
• Bramley R.G.V., White R.E. The effect of pH, liming, moisture and temperature on the activity of nitrifiers in a soil under pasture. Aust. J. Soil Res. 1989;27:711-724.
• Bussiere F., Cellier P. Modification of the soil temperature and water content regimes by a crop residue mulch: Experiment and modelling. Agric. For. Meteorol. 1993;68:1-28.
• Cabidoche Y.M. Management of soil fertility in the Caribbean zone. (In French). Conjonction 1996;200:122-129.
• Cabrera M.L., Kissel D.E. Potentially mineralizable nitrogen in disturbed and undisturbed soil samples. Soil Sci. Soc. Am. J. 1988;52:1010-1015 a.[Abstract/Free Full Text]
• Cabrera M.L., Kissel D.E. Evaluation of a method to predict nitrogen mineralized from soil organic matter under field conditions. Soil Sci. Soc. Am. J. 1988;52:1027-1031 b.[Abstract/Free Full Text]
• Chotte J.L., Feller C., Vallony M.J., Nicolardot B. Disponibilité de l'azote dans les sols cultivés des Petites Antilles. (In French, with English abstract.) Agric. Développement 1994;4:23-30.
• Deans J.R., Molina J.A.E., Clapp C.E. Models for predicting potentially mineralizable nitrogen and decomposition rate constants. Soil Sci. Soc. Am. J. 1986;50:323-326.[Abstract/Free Full Text]
• Dommergues Y., Mangenot F. Ecologie microbienne du sol. Paris: Masson et Cie, 1970.
• Ellert B.H., Bettany J.R. Temperature dependence of net nitrogen and sulfur mineralization. Soil Sci. Soc. Am. J. 1992;56:1133-1141.[Abstract/Free Full Text]
• Greenland D.J., Wild A., Adams D. Organic matter dynamics in soils of the tropics—From myth to complex reality. In: Lal R., Sanchez P.A., eds. Myths and science of soils of the tropics. Madison, WI: SSSA and ASA, 1992:17-33 SSSA Spec. Publ. 29..
• Grissi B., Grace C., Brookes P.C., Benedetti A., Dell'Abate M.T. Temperature effects on organic matter and microbial biomass dynamics in temperate and tropical soils. Soil Biol Biochem. 1998;30:1309-1315.
• Hankinson T.R., Schmidt E.L. An acidophilic and a neutrophilic Nitrobacter strain isolated from the numerically predominant nitrite-oxidizing population of an acid forest soil. Appl. Environ. Microbiol. 1988;54:1536-1540.[Abstract/Free Full Text]
• Hart S.C., Stark J.M., Davidson E.A., Firestone M.K. Nitrogen mineralization, immobilization and nitrification. In: Weaver R.W., et al. , ed. Methods of soil analysis. Part 2. SSSA Book Ser. 5. Madison, WI: SSSA, 1994:985-1018.
• Ingham E.R., Horton K.A. Bacterial, fungal and protozoan responses to chloroform fumigation in stored soil. Soil Biol. Biochem. 1987;19:545-550.
• INRA. 1996. Climatic abstract: Measurements realized from 1986 to 1996 at the Domaine Duclos (Guadeloupe). (In French.) INRA, Station Agropédoclimatique de la Zone Caraïbe, Guadeloupe.
• Joergensen R.G., Brookes P.C., Jenkinson D.S. Survival of the soil microbial biomass at elevated temperatures. Soil Biol. Biochem. 1990;22:1129-1136.
• Kampshake L.J., Hannah S.A., Cohen J.M. Automated analysis for nitrate by hydrazine reduction. Water Resour. Res. 1967;1:205-216.
• Kaplan A. Standard methods of clinical chemistry. New York: Academic Press, 1965.
• Lal R., Ragland J. Agricultural sustainability in the tropics. In: Ragland J., Lal R., eds. Technologies for sustainable agriculture in the tropics. Madison, WI: ASA, CSSA, and SSSA, 1993:1-6 ASA Spec. Publ. 56..
• Luizao F., Luizao R., Chauvel A. Premiers résultats sur la dynamique des biomasses racinaires et microbiennes dans un latosol d'Amazonie centrale (Brésil) sous forêt et sous patûrage. (In French, with English abstract.) Cah. Orstom 1992;27:69-79.
• Mary B., Recous S. Measurement of nitrogen mineralization and immobilization fluxes in soil as a means of predicting net mineralization. Eur. J. Agron. 1994;3:291-300.
• Mench M., Clairon M. Observations préliminaires sur l'évolution de quelques propriétés d'un sol ferrallitique de Guadeloupe (FWI) après un apport de boues urbaines. (In French, with English summary). Agronomie (Paris) 1991;11:283-291.
• Myers R.J.K. Temperature effects on ammonification and nitrification in a tropical soil. Soil Biol. Biochem. 1975;7:83-86.
• Nordmeyer H., Richter J. Incubation experiments on nitrogen mineralization in loess and sandy soils. Plant Soil 1985;83:433-445.
• Prior P., Beramis M. Induction de la résistance au flétrissement bactérien dû à Pseudomonas solanacearum E F Smith chez un cultivar de tomate réputé sensible. (In French, with English summary). Agronomie (Paris) 1990;10:391-401.
• Sanchez P.A., Logan T.J. Myths and science about the chemistry and fertility of soils in the tropics. In: Lal R., Sanchez P.A., eds. Myths and science of soils of the tropics. Madison, WI: SSSA and ASA, 1992:35-46 SSSA Spec. Publ. 29..
• Sanchez, P.A., C. Palm, L.T. Szott, E. Cuevas, and R. Lal. 1989. Organic input management in tropical agroecosystems. p. 125–242. In D.C. Coleman, et al. (ed.) Dynamics of soil organic matter in tropical ecosystems. NifTAL Project, College of Tropical Agriculture and Human Resources, Manoa, Hawaii.
• Schmidt E.L., Belser L.W. Autotrophic nitrifying bacteria. In: Weaver R.W., et al. , ed. Methods of soil analysis. Part 2. Madison, WI: SSSA, 1994:159-177 SSSA Book Ser. 5..
• Sierra J. N mineralization and its error of estimation under field conditions related to the light fraction soil organic matter. Aust. J. Soil Res. 1996;34:755-767.
• Sierra J. Temperature and soil moisture dependence of N mineralization in intact soil cores. Soil Biol. Biochem. 1997;29:1557-1563.
• Stanford G., Frere M.H., Schwaninger D.H. Temperature coefficient of soil nitrogen mineralization. Soil Sci. 1973;115:321-323.
• Stanford G., Smith S.J. Nitrogen mineralization potentials of soils. Soil Sci. Soc. Am. Proc. 1972;36:465-472.
• Stockdale E.A., Rees R.M. Release of nitrogen from plant and animal residues and consequent plant uptake efficiency. In: Kristensen L., et al. , ed. Nitrogen leaching in ecological agriculture. Bicester, UK: AB Academic Publishers, 1995:229-245 Proc. Int. Workshop, Royal Veterinary and Agricultural University..
• Toyota K., Ritz K., Young I.M. Survival of bacterial and fungal populations following chloroform-fumigation: Effects of soil matric potential and bulk density. Soil Biol Biochem. 1996;28:1545-1547.
• Vance E.D., Brookes P.C., Jenkinson D.S. Microbial biomass measurements in forest soils: determination of kC values and tests of hypotheses to explain the failure of the chloroform fumigation–incubation method in acid soils. Soil Biol. Biochem. 1987;19:689-696.

This article has been cited by other articles:

				A. Suman, M. Lal, A. K. Singh, and A. Gaur Microbial Biomass Turnover in Indian Subtropical Soils under Different Sugarcane Intercropping Systems Agron. J., May 3, 2006; 98(3): 698 - 704. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (16) Citing Articles via Google Scholar Google Scholar Articles by Sierra, J. Articles by Marbán, L. Search for Related Content PubMed Articles by Sierra, J. Articles by Marbán, L. GeoRef GeoRef Citation Agricola Articles by Sierra, J. Articles by Marbán, L.