| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Natural Resource Management, Heinrich-Buff-Ring 26-32, Justus-Liebig University Gießen, D-35392 Gießen, Germany
b Fraunhofer Institute for Atmospheric Environmental Research (IFU), Div. Biosphere/Atmosphere Exchange, Dep. of Soil Microbiology, Kreuzeckbahnstr.19, D-82467 Garmisch-Partenkirchen, Germany
* Corresponding author (butterbach{at}ifu.fhg.de)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: BaPS, Barometric Process Separation • ECD, electron capture detector • GCMS, gas chromatograph-mass spectrometer • RC, respiration coefficient • TCD, thermal conductivity detector • WFPS, water-filled pore space
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Most of these problems can be overcome by using a new approach, the BaPS (Ingwersen et al., 1999), which allows the determination of gross N-nitrification rates in oxic soils without destroying the original structure of the soil or having to add labeled substrates to the soil. With this method, gross nitrification rates are calculated from the balances of O2, CO2, and the total gas amount in a closed, isothermal system. The BaPS technique allows a high temporal resolution of measurements, and thus, it is useful for analysis of the dependency of gross nitrification rates on variables such as temperature or soil moisture. The latter point, which has not been investigated so far in detail, maybe of crucial importance to understand seasonal changes in the gross turnover of N in a given ecosystem. Some work has been done in temperate ecosystems by Davidson et al. (1990)(1992) and Stark and Hart (1997). With regard to tropical forest ecosystems, detailed information about gross nitrification rates is extremely scarce (Zou et al., 1992; Neill et al., 1999; Riley and Vitousek, 1995) and no information about the effects on nitrification of soil temperature and soil moisture exists. Within a research program on factors controlling N2O-emissions from tropical forest ecosystems in Australia (Breuer et al., 2000) an extensive study on the effect of changes in soil temperature and soil moisture on gross nitrification rates and on N2O-production was performed by using the BaPS technique.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
40 km southwest of Cairns, the Atherton Tablelands constitute an elevated plateau with an average altitude of 850 m above sea level (17°09' S to 17°37' S lat. and 145°25' E to 145°45' E long.). Soil parent materials of the sites investigated were granite (Kauri Creek), metamorphics (Lake Eacham; mainly schist and phyllit), and acid volcanics (Massey Creek; mainly rhyolite) (Laffan, 1988). Details about soil characteristics, mean annual temperatures, and mean annual precipitation for the three study sites are given in Table 1 (see also Breuer et al., 2000).
|
For the investigation of seasonal changes in gross nitrification rates, intact soil cores were taken during the different hygric seasons at the Atherton Tablelands. Intact cores were collected in August 1997 during the dry season, and cores representing the wet season were collected in March 1998. A third set of cores was collected during the transition period from dry to wet season in January 1999 (see also Breuer et al., 2000).
Determination of Gross Nitrification Rates Using Barometric Process Separation
Gross nitrification rates were determined using the BaPS technique. Theoretical considerations, mathematical procedures of the BaPS technique and a comparison of results obtained for gross nitrification rates determined with both the 15N isotope pool-dilution technique as well as the BaPS technique were described in detail by Ingwersen et al. (1999). As BaPS is a new approach to determine gross nitrification rates in soils, a short concept of the method is given. Barometric process separation is based on the determination of the CO2, O2, and total gas balances inside an isothermal, gas tight soil system. Nitrification (net consumption of O2 by the ammonia-oxygenase - pressure decrease), denitrification (net CO2 production and net N-gas production, i.e., NO, N2O, and N2, here referred to as NxOy - pressure increase) and soil respiration (pressure neutral, if coefficient of respiration [RC] is equal to 1.0) are the main biological processes responsible for gas pressure changes inside such a system. Therefore, the total pressure change (
n/t) inside the system consists of the net changes of O2 (O2/t), CO2 (CO2/t), and gaseous N-compounds via denitrification (NxOy/t):
 | [1] |
Since the total pressure change (n/t) and the net changes of O2 (O2/t) and CO2 (CO2/t) can be measured directly, the production of gaseous N-compounds via denitrification (NxOy/t) can be calculated. Inverse balancing of the total gas balance leads to gross nitrification rates (Ingwersen et al., 1999). Since the calculation of gross nitrification rates by the BaPS technique is sensitive to deviations of RC from the ideal value of 1.0, we determined in a first step the RC by solving the gas balances for CO2 and O2. If the calculated RC is not equal to 1.0 this changes the basic assumption for determination of gross nitrification rates with BaPS (see Ingwersen et al., 1999). Therefore, an iterative approach was used, in which the initially calculated RC value was used to solve the BaPS-equation system again. This procedure was repeated until the inserted RC value equaled the calculated RC. In most cases, the RC leveled off at values between 0.95 and 1.05. Experiments were rejected if these values were over- or underestimated (<10% of all cases), since the error in calculating gross nitrification by BaPS might become too large (Ingwersen et al., 1999).
Since information about the importance of heterotrophic versus autotrophic nitrification was not available for the tropical rain forest sites in Australia, a 1:1 ratio between autotrophic and heterotrophic nitrification was supposed since first results from measurements of cell numbers of autotrophic and heterotrophic nitrifiers suggest that both microorganism groups are present in these soils. In consequence of considering autotrophic nitrification within the BaPS-equation system, the CO2 assimilation by autotrophic nitrifiers must be considered as well (for equations see Ingwersen et al., 1999). It must be stressed that neglecting CO2 assimilation during autotrophic nitrification has only a minor bias on the final values obtained for nitrification rates, denitrification rates, and soil respiration rates (<14%, Ingwersen et al., 1999). However, by assuming a 1:1 ratio between autotrophic and heterotrophic nitrification and, thus, by considering the CO2 assimilation during autotrophic nitrification, the possible error of neglecting autotrophic nitrification processes in soils will be cut in half (<7%).
Incubation of Soil Cores and Determination of Oxygen and Carbon Dioxide Concentrations
Three intact soil cores were incubated in parallel in a custom-made isothermal water bath, which was controlled by a thermostat (N8-C41, Haake, Karlsruhe, Germany) (Fig. 1)
. All walls of the water bath were darkened with aluminum foil and additionally insulated by 30-mm-thick polystyrene foam panels. Prior to experiments, gas tightness of the incubation cylinders containing the soil cores was tested, as described by Ingwersen et al. (1999). To insure that if the soil cores were equilibrated to the water-bath temperature, an additional temperature sensor, connected to a digital thermometer GTH 1200 A (Greisinger, Regenstauf, Germany), was inserted in the system and experiments were started when the difference between inside and outside temperature was <0.2°C. The air pressure inside the closed incubation system was recorded continuously by pressure sensors (range: 800–1100 hPa, sensitivity 0.1%, GMDP, Greisinger, Regenstauf, Germany) and data were recorded on a PC (Fig. 1) using a data acquisition system. Carbon dioxide and O2 concentrations within the system were determined at the beginning and end of each experiment by withdrawing head space samples with a gas tight syringe via a septum (Fig. 1), analyzing for CO2 with a thermal conductivity detector (TCD), and for O2 with a 63Ni-electron capture detector (ECD) (Perkin Elmer, Uberlingen, Germany). Specifications for gas chromatography were: carrier gas He 5.0 15 mL min-1 (TCD) and 30 mL min-1 (ECD); purge gas 50 mL L-1 CH4 in Ar at a flow rate of 40 mL min-1 (ECD); detector temperature 250°C (TCD) and 350°C (ECD); analytical column GS-Q Megabore with split end connected to TCD and ECD (30 m length, 
0.53 mm; J&W Scientific, Folsom, CA). The ECD and TCD were calibrated regularly with standard gas: 151 mL O2 L-1 in N2 5.0 and 50 mL CO2 L-1 in synthetic air (all gases by Messer-Griesheim, Olching, Germany). To minimize the error in the determination of O2 and CO2 concentrations, measurements were done at least in six-fold replicates. For the acquisition and analysis of chromatography data a CSI interface (Nelson Analytical, Cubertino, CA), and the APEX 2.15 software (Flowchem, Besigheim, Germany) were used. At the end of all experiments, the gravimetric water content was determined for each soil core.
|
Experiments with Nitrification Inhibitors
To further validate, the BaPS-methodology experiments were carried out in which the magnitude of the nitrification rate in the soil was influenced by the use of the nitrification inhibitors 2-chloro-6-(trichloromethyl)-pyridine (nitrapyrin) (McCarty and Bremner, 1989; Regina et al., 1998) and acetylene (Yoshinari et al., 1977; Davidson et al., 1993). In the first set of experiments nitrapyrin was added to sieved soil from the Kauri Creek site, and rates of gross nitrification in the soil as determined by the BaPS technique before and after addition of nitrapyrin were compared (incubation temperature: 20°C; soil moisture: 21.5; N = 3). In the second set of experiments rates of gross nitrification were measured before and after addition of 0.1% of C2H2 to the head space of the BaPS incubation system (incubation temperature: 20°C; soil moisture: 21.5; N = 3). In both experiments total incubation time was 36 h.
Determination of Nitrous Oxide Emission Rates from Intact Soil Cores
To study if gross nitrification rates in the soil cores can be directly related to the emission of the radiative active trace gas, N2O, from these soils, additional measurements of N2O-emissions were performed. For these experiments soil cores were sealed gas-tight for 1 h, 3 h prior to measurements of gross nitrification rates via BaPS and the increase of N2O concentration in the head space of the system was followed by analyzing gas samples for N2O by gas chromatography as described above (calibration standards used: 0.6 µL N2O L-1 and 15 µL N2O L-1 in synthetic air; Messer-Griesheim, Olching, Germany). Nitrous oxide-emission rates were calculated from the linear increase of N2O concentration inside the sealed chambers. All flux rates were corrected for temperature and air pressure. The sensitivity for measurements of N2O at ambient atmospheric N2O background concentration (313 nL L-1) was 10 nL L-1. Hence, the detection limit for N2O emissions from intact soil cores at a given head-space height of 0.1 m (head-space volume 1.131 L) was 1.3 µg N2O-N m-2 h-1.
Determination of the Dependency of Nitrification and Nitrous Oxide Emission from Soil Temperature and Soil Moisture
To parameterize the temperature sensitivity of nitrification and N2O-emission in tropical-forest soils, the soil cores were incubated at five different temperatures within 1 wk. The temperature range chosen for the incubation experiments (14–23°C) corresponds to the annual range of air temperatures as observed at the Atherton meteorological station (Laffan, 1988) or during the field trip to Australia (Breuer et al., 2000). Experiments were conducted in 2°C steps, randomly distributed between 14 and 22°C (experiments in 1997) or 15° through 23°C (experiments in 1998).
The effect of soil moisture on rates of nitrification and N2O emission was studied by simulating precipitation events, which, in steps of 1 mm, added up to 5 mm of precipitation (Exp. 1) or in a second set of experiments, using steps of 3, 4, and 5 mm (two additions), added up to 17 mm of precipitation. Precipitation was simulated, by dripping a standard rain mixture (11 mg CaCl2, 24.4 mg KCl, 18.6 mg Na2SO4 in 1000 mL of distilled water; Schierl, 1991) randomly onto the soil surface via syringe. After each moisture addition, the soil was allowed to equilibrate to the new conditions for 12 h before measurements of gross nitrification and N2O emission started. Approximately 1 wk was needed for these experiments, i.e., to analyze nitrification and N2O emission starting at field soil moisture conditions up to the soil moisture conditions achieved after addition of 5 or 17 mm of rainfall, respectively. At the end of experiments, gravimetric soil moisture contents were measured for all soil cores. Water-filled pore space was calculated on the basis of soil volume, SDW, bulk density, particle density, pore space, and water content. In contrast to Linn and Doran (1984), we used a particle density of 1.8 g cm-3 to calculate WFPS. This was done, since the soil in the cores includes both, the litter layer (organic material density: 1.4 g cm-3) and the mineral soil (mineral soil density: 2.65 cm-3). The portion of organic material in the individual cores was in total 50 to 70%.
Statistical Analyses
All statistical analyses were performed with SPSS 8.0 (SPSS, Chicago, IL) and Microcal Origin 4.0 (Microcal Software, Northampton, MA).
Because of nonnormal distribution of nitrification rates, the nonparametric Mann-Whitney-test was used (i) to identify differences between soil cores from the three research sites and (ii) to compare seasonal means of gross nitrification rates. In addition to linear regression analysis, exponential, polynomial, and logarithmic curve fitting procedures were tested in addition. Correlation and regression analyses were declared as significant if P < 0.05.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Application of nitrification inhibitors to soil samples taken from the Kauri Creek site resulted in a strong, significant inhibition (P < 0.05) of gross nitrification rates. Mean gross nitrification in the untreated soil samples was 3.31 ± 0.2 µg N g-1 SDW d-1. After addition of nitrapyrin, gross nitrification dropped down to 2.01 ± 0.1 µg N g-1 SDW d-1, i.e., a reduction of nitrification 39.4%. Comparable results were also obtained when 0.1% of acetylene was added to the head space of the incubation system (mean gross nitrification rate 1.96 ± 0.1 µg N g-1 SDW day-1, reduction: 40.1%). The observation, that nitrification inhibitors do not fully inhibit nitrification in soils is not new. Mosier (1980), Regina et al. (1998), Garrido et al. (2000), and others demonstrated that the inhibitory effect of low concentrations of acetylene or nitrapyrin on nitrification depends for example on the soil type and that the inhibition is incomplete (acetylene: 0–92%; nitrapyrin: 45–96%; McCarty and Bremner, 1989), whereas other groups found that even low concentrations of acetylene inhibited nitrification by 100% (e.g., Bremner and Blackmer, 1979; Bollmann and Conrad, 1997). However, all these investigations have not directly studied if gross nitrification was inhibited, but have used other indicators like net nitrification, changes in soil nitrate or nitrite concentrations, or N2O- and NO-production as indicators. Pedersen et al. (1999) studied rates of gross nitrification in forest soils taken from the northern Sierra Nevada Mountains and found that gross nitrification rates were inhibited by acetylene by 35.8% (average of all sites investigated; range: 8–81%), a figure very close to the one we found for tropical-forest soils in Australia. Furthermore, the authors pointed out that the remaining nitrification activity must have been because of heterotrophic nitrification—an explanation which is also most likely for our soils, since it is well known that some heterotrophic nitrifiers cannot be inhibited by acetylene or other nitrification inhibitors (e.g., Schimel et al., 1984; Stams et al., 1990).
Temperature Dependency of Gross Nitrification Rates
Nitrification rates in soil cores taken from the three tropical rain-forest sites in Australia during the dry season in 1997 showed a strong dependency on changes in temperature; i.e., nitrification increased with increasing soil temperature (Fig. 2) . For all investigated sites, WFPS was 32 (1% across all incubation temperatures, the highest rates of nitrification were found for the Kauri Creek site (7.8 ± 0.4 [14°C] to 20.0 ± 1.0 mg NH+4-N m-2 h-1 [22°C], Fig. 2). Linear regression analysis confirmed the close relationship between temperature and gross nitrification at the Kauri Creek site (nitrification [mg NH+4-N m-2 h-1] = -13.2 + 1.49 x soil temperature, r2 = 0.96, P = 0.003). Nitrous-oxide emissions from soil cores taken at the Kauri Creek site ranged from 240 to 310 µg N2O-N m-2 h-1, but a significant temperature effect on N2O emissions was not observed (Fig. 2). Nitrification rates at the Lake Eacham site were 25% of those at the Kauri Creek site and no significant correlation between temperature and gross nitrification rates or N2O emissions was observed, although highest rates for N2O emissions (685 µg N2O-N m-2 h-1) and gross nitrification were found at temperatures of 20 and 22°C, respectively (Fig. 2). Gross nitrification rates as well as N2O emissions at Massey Creek site were both positively correlated with increases in soil temperature (Fig. 2), but rates of nitrification were 30% lower as compared with the Kauri Creek site. The linear regression between temperature and nitrification rates can be described by the equation:
 | [2] |
|
 | [3] |
Rates of gross nitrification in soil cores taken from the sites at Kauri Creek, Massey Creek and Lake Eacham during the wet season in 1998 also showed in most cases a positive relationship with soil temperature (Fig. 3)
. An average WFPS of 34 ± 2 % for all research sites was slightly higher compared with soil moisture conditions in 1997. Nitrification rates at the Kauri Creek site were lower than rates of nitrification as observed for soil cores taken during the dry season in 1997. Even though a tendency for increasing rates of nitrification with increasing temperature is observed, the linear regression was not significant (P = 0.25). Nitrogen-dioxide emissions from the intact soil cores were 20% of those in 1997 (mean 1998: 52.9 ± 9.6 µg N2O-N m-2 h-1, mean 1997: 308.8 ± 41.8 µg N2O-N m-2 h-1) (Fig. 2 and 3), and, as in 1997, a significant correlation between temperature and N2O emission was not found. Because of the high value for gross nitrification at the Lake Eacham site at 15°C, the correlation of r2 = 0.69 between gross nitrification and temperature was only significant with P = 0.081. The magnitude of nitrification at the Lake Eacham site was comparable with the range of nitrification rates observed in 1997 for this site (Fig. 2 and 3), but, as already mentioned for the Kauri Creek site, N2O emissions were significantly lower for soil cores taken during the wet season in 1998 (mean: 9.5 ± 1.2 µg N2O-N m-2 h-1) than for soil cores taken during the dry season in 1997 (mean: 415.8 ± 92.8 µg N2O-N m-2 h-1). Nitrification rates (up to 23.2 ± 5.3 mg NH+4-N m-2 h-1) for the Massey Creek site were 1.5-fold greater for soil cores taken in 1998 as compared with soil cores taken in 1997. At this site, as in 1997, a strong correlation of nitrification rates with temperature was found (nitrification [mg NH+4-N m-2 h-1] = -31.4 + 2.39 x soil temperature, r2 = 0.98, P = 0.001). However, a significant correlation could not be demonstrated (in contrast to measurements in 1997) for N2O emissions (mean 1998: 25.5 ± 3.8 µg N2O-N m-2 h-1; mean 1997: 461.8 ± 91.9 µg N2O-N m-2 h-1).
|
|
|
In Exp. 1, gross nitrification rates ranged from 13.3 to 36.5 mg NH+4-N m-2 h-1 and N2O emissions ranged from 3.6 to 80.4 µg N2O-N m-2 h-1 (Fig. 5)
. Two different trends in gross nitrification were observed in response to simulated rainfall with concomitant increases in WFPS: both, the Kauri Creek and the Massey Creek sites, showed decreasing nitrification rates, but rising N2O emissions with increasing soil moisture (increase of WFPS of 5% for both sites), whereas for the Lake Eacham site, a low amount of precipitation (increase of WFPS 4.0%) activated nitrification, but did not change N2O emissions (Fig. 5).
|
|
60%. In our experiments on undisturbed soil cores, gross nitrification rates at these values of WFPS were severely reduced (Massey Creek), or fell below the limit of detection (Kauri Creek and Lake Eacham). A possible explanation for this might be the differences in the soil types investigated: mineral soil samples with a generally lower pore space and bulk density in the case of Linn and Doran (1984) and soils containing higher proportions of organic matter, low bulk densities and high pore spaces as were sampled in our experiments. The observation that nitrification rates decreased with increasing soil moisture is in part contrary to findings of Ingwersen et al. (1999). They found that gross nitrification rates for intact soil cores taken from a temperate spruce forest were highest when the water-holding capacity of the mineral soil was reached. However, since the soil cores in those experiments comprised the mineral soil as well as the thick (up to 8 cm) forest floor, in which most of the microbial activity can be found (Papen and Butterbach-Bahl, 1999; Menyailo and Huwe, 1999), the authors concluded, that even under waterlogged conditions in the mineral layer, high gross nitrification rates can occur in the aerobic organic layer. Such a thick organic layer did not exist at the research sites in Australia. We conclude that the observation of decreasing nitrification rates with strongly increasing values of soil moisture is most likely associated with the appearance or extension of anaerobic microsites because of restricted O2 diffusion into the soil at higher soil moisture contents (e.g., Tiedje et al., 1984).
Site and Seasonal Variations in Gross Nitrification
To identify differences in nitrification activity between sites and between different hygric seasons at a given site, only those measurements of nitrification were evaluated that had been performed when (i) incubation temperatures were comparable with the mean soil temperatures as observed in the field during the time when soil cores were taken (04–07/97, 14–18°C; 02–03/98, 19–23°C; 11/98–01/99, 23°C) and field N2O-emission measurements were carried out (Breuer et al., 2000) and when (ii) incubation soil moisture conditions were field fresh (i.e., before water was added to the soil cores to simulate the effect of precipitation events on nitrification).
Table 3 shows that for the Kauri Creek site no significant differences between nitrification rates in the dry season and the wet season could be demonstrated, whereas nitrification rates in the transition period from dry to wet period were significantly enhanced (overall maximum of 27.6 ± 6.1 mg NH+4-N m-2 h-1). The latter finding, i.e., highest rates of nitrification during the transition period from dry to wet season, was also observable at Lake Eacham (14.8 ± 5.1 mg NH+4-N m-2 h-1) and Massey Creek site (26.0 ± 3.9 mg NH+4-N m-2 h-1). For both of these sites, nitrification rates were also higher during the wet season as compared with the dry season (Table 3). The fact that nitrification rates in the transition period from dry to wet season are enhanced can be explained by the strong accumulation of litter during the dry season—which may accumulate to a height of up to 15 cm on the floor (Mike Hopkins, personal communication, 1997)—and its mineralization afterwards within few weeks at the beginning of the wet season. Therefore, since substrate is available and soil moisture is beneficial during this period, high rates of nitrification are very plausible. The same observation, i.e., very high nitrification rates after artificially rewetting soils from drought-stressed deciduous forests in Mexico at the end of the dry season, was also made by Davidson et al. (1993).
|
Correlation Between Gross Nitrification in Intact Soil Cores and in situ Nitrous Oxide Fluxes
A correlation analysis was performed to determine whether a link between mean N2O fluxes as measured in the field (Breuer et al., 2000) and mean nitrification rates as measured on intact soil cores directly at the end of the different field measuring campaigns exist. Figure 7
shows a positive, though nonsignificant correlation between gross nitrification and N2O emissions (N2O emission [µg N2O-N m-2 h-1] = 13.7 + 0.0019 x nitrification rate [µg NH+4-N m-2 h-1]; r2 = 0.257; P = 0.163). If we exclude the outlier at the Kauri Creek site, March 1998, a strong positive linear correlation between nitrification rates and N2O emissions in the field was observed (N2O emission [µg N2O-N m-2 h-1] = 0.9 + 0.0022 x nitrification rate [µg NH+4-N m-2 h-1]; r2 = 0.750; P = 0.005). One reason for the outlier at the Kauri Creek site (03/98) and the differing ratio of N2O-flux rate and gross nitrification rate may be the time lag between measurements of N2O fluxes in the field and determination of gross nitrification rates in the laboratory in Germany. Measurements of N2O emissions at Kauri Creek (03/98) were carried out during the height of wet season (Breuer et al., 2000) whereas the soil cores for determination of gross nitrification rates were collected 3 wk later at the end of wet season. Furthermore, because of the high microbial activity under favorable conditions at the peak of the wet season (high temperature and high soil moisture), substrate availability at the end of the wet season may have been reduced, leading to overall lower N-turnover rates, and hence reduced gross nitrification rates. Therefore, measurements of nitrification rates and N2O-emissions at the same time in the field are necessary to unequivocally demonstrate this relationship.
|
20-fold higher than the N2O losses during nitrification estimated by Ingwersen et al. (1999) for a temperate forest soil (0.008–0.053%). Using the data published by Davidson et al. (1993) a ratio of N2O-N to gross nitrification of 0.05% can be calculated; a value laying between the ranges given above.
Effect of Climate and Site Management on Gross Nitrification Rates in Soils
As shown in our experiments, gross nitrification rates are highly dependent on soil moisture and temperature conditions. The influence of these two climate-driven variables is reflected in the geographical pattern of the strength of gross nitrification rates reported so far. Average nitrification rates in the generally moist, warm soils of tropical forest ecosystems (160 ± 45 mg NH+4-N m-2 h-1) are 50 to 60% higher than those found in forest soils in temperate (114 ± 21 mg NH+4-N m-2 h-1) or etesian (Mediterranean) (104 ± 22 mg NH+4-N m-2 h-1) climates (Table 2). Although, because of the high variability in the reported data, no statistically significant differences between these geographical regions can be confirmed (Mann-Whitney-Test P = 0.859 for tropical vs. etesian; P = 0.762 for tropical vs. temperate; P = 0.570 etesian vs. temperate) (Table 2). Nitrification rates presented here for the tropical rain forest ecosystems in Australia and data from Zou et al. (1992) (tropical rain forests of Costa Rica) and Neill et al. (1999) (tropical rain forests of the Amazon Basin, Brazil) range from 46 to 662 mg NH+4-N m-2 h-1, which is much higher than nitrification rates found in younger soils of different geological age in montane rain forests of Hawaii (Riley and Vitousek, 1995) (range 19–114 mg NH+4-N m-2 h-1). The reason for this discrepancy may be the geological age of the soils in Hawaii, with most of them derived from younger parent materials. In the chronosequence that Riley and Vitousek (1995) investigated, they found the highest rates of nitrification in the oldest soil (4.5 x 106 yr). Ecosystems evolved over long time periods on old substrates, have had time for N accumulation via N fixation, and other inputs. Some tropical forests growing on old substrates tend to have higher N-pool sizes like organic N and NH+4 (Vitousek and Sanford, 1986) and more rapid rates of N cycling (Matson and Vitousek, 1987; Livingston et al., 1988), leading to higher gross nitrification rates. Disregarding these Hawaiian soils increases the value of average gross nitrification rates to 271 ± 52 mg NH+4-N m-2 h-1 for tropical soils, and yields a significant difference from temperate and etesian regions (Mann-Whitney-Test P = 0.009 and P = 0.013 respectively). The work of Neill et al. (1999) suggests that the obviously high rates of N turnover in tropical forest ecosystems are very sensitive to land-use changes, since nitrification rates are strongly reduced if tropical forests are converted to pasture.
Information on gross nitrification rates in subtropical dry regions is scarce. To our knowledge only Kaye and Hart (1998) published values for untreated coniferous forest soils in Arizona (Table 2). Davidson et al. (1993) conducted measurements on soil cores at the end of a dry season in Mexico. Nitrification increased dramatically up to values of over 2000 mg NH+4-N m-2 h-1 upon wetting of the soil, so that the mean residence time of N in inorganic pools of wetted soil was on the order of 1 to 2 d. This is in good agreement with results we obtained for gross nitrification rates of up to 662 mg NH4 + N m2 h-1 in the transition period towards the wet season when substrate availability is high because of the prolonged accumulation of organic matter in dry season. We speculate that presumed N-substrate availability is sufficient and that nitrification in warmer and dryer climates is limited by low soil moisture, whereas in temperate regions nitrification may be more limited by lower temperatures. This assumption is supported by data published by Stark and Hart (1997). Using their data on gross nitrification in 11 different forest stands in a multiple regression analysis, the influence of mean annual precipitation (MAP [mm]), nitrate (NO-3-N [mg N kg-1]), and elevation (ELEV [meters above sea level]; reflecting temperature and site conditions) can be expressed as: nitrification [mg NH+4-N m-2 d-1] = 83.8 + 0.0424 x MAP + 71.9·NO-3-2.94 x ELEV (corrected r2 = 0.78; P = 0.015; Durbin-Watson-Test = 1.51; Tolerance = 0.636–0.935). Standardized Beta-coefficients reveal that elevation (-0.450) and precipitation (0.415) are the main driving parameters compared with NO-3 (0.348). Well-controlled, factorial-design experiments with soils from different ecosystems and with a specified set of varying parameters should prove useful in elucidating the effects of environmental variables on nitrification across differing climates.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
Received for publication November 10, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Menendez, P. Merino, A. Lekuona, M. Pinto, C. Gonzalez-Murua, and J. M. Estavillo The Effect of Cattle Slurry Electroflotation Products as Fertilizers on Gaseous Emissions and Grassland Yield J. Environ. Qual., May 1, 2008; 37(3): 956 - 962. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Ingwersen, U. Schwarz, C. F. Stange, X. Ju, and T. Streck Shortcomings in the Commercialized Barometric Process Separation Measuring System Soil Sci. Soc. Am. J., January 11, 2008; 72(1): 135 - 142. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. J. Livesley, M. A. Adams, and P. F. Grierson Soil Water Nitrate and Ammonium Dynamics under a Sewage Effluent Irrigated Eucalypt Plantation J. Environ. Qual., October 24, 2007; 36(6): 1883 - 1894. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Agehara and D. D. Warncke Soil Moisture and Temperature Effects on Nitrogen Release from Organic Nitrogen Sources Soil Sci. Soc. Am. J., September 29, 2005; 69(6): 1844 - 1855. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Muller, M. K. Abbasi, C. Kammann, T. J. Clough, R. R. Sherlock, R. J. Stevens, and H.-J. Jager Soil Respiratory Quotient Determined via Barometric Process Separation Combined with Nitrogen-15 Labeling Soil Sci. Soc. Am. J., September 1, 2004; 68(5): 1610 - 1615. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
|
