| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Lab. of Soil Ecology and Microbiology, 024 Coastal Institute in Kingston, Univ. of Rhode Island, Kingston, RI 02881
b Dep. of Crop, Soil, and Environmental Sciences, 115 Plant Science Building, Univ. of Arkansas, Fayetteville, AR 72701
* Corresponding author (jam7740u{at}postoffice.uri.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: ARTF, artificial burrows • CTRL, control • DOC, dissolved organic C • LEAF, artificial burrows containing corn leaves • WORM, amended with earthworms
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
We conducted a mesocosm-scale experiment under laboratory conditions to evaluate the dynamics of C and N in the soil surrounding burrows created by the activities of earthworms. We were particularly interested in discerning the contributions of litter enrichment, macropores, and earthworms to C and N dynamics in drilosphere soil. Studies aimed at elucidating the mechanisms by which anecic earthworms affect nutrient transformation in soil often make no distinction between burrow and bulk soil. However, the effects of anecic earthworms on soil tend to be highly localized (e.g., Binet and Curmi, 1992; Bohlen et al., 1997; Devliegher and Verstraete, 1997; Görres et al., 1997, 2001), such that they may be masked by mixing of burrow soil with adjacent bulk soil (Görres et al., 1997). As such, our study focused on effects on burrow soil. The main objectives of the study were to: (i) assess the relationship between dynamics of corn litter consumption by L. terrestris and C and N dynamics in burrow soil, and (ii) evaluate the extent to which macropores and litter incorporation contribute to the effects of anecic earthworms on C and N dynamics in burrow soil. Our research question was: Is the worm necessary, or can litter enrichment and enhanced movement of gases and water in macropores account for the effects of earthworms on burrow soil C and N dynamics?
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Plant Litter and Earthworms
Corn leaves were collected from a no-till, organic cornfield at Casey Farm in Saunderstown, RI in October of 1998. Leaves were stored in plastic bags in the dark at 4°C for approximately 4 mo to allow for partial decomposition of the plant material. Leaves were cut into approximately 2-cm long pieces on the day the experiment was begun. The mean C concentration of the litter was 380 g kg-1, with an N concentration of 18 g kg-1 and it had a C/N ratio of 21:1, determined using a Carlo Erba N/C analyzer (model NA 1500, series 2, Carlo Erba, Milan, Italy).
Earthworms were purchased from a commercial outlet in North Kingstown, RI 1 d before the beginning of the experiment and stored in the dark at 4°C. The fresh weight of the worms was determined based on measurements of 24 individuals the day the experiment was started.
Mesocosms
Experimental mesocosms consisted of 10-cm i.d. and 0.5-m long white polyvinyl chloride (PVC) cores filled with sieved soil to a depth of 40 cm. The bottom and top of the core were fitted with a fine fiberglass mesh (held in place with a rubber band) to retain soil and prevent earthworm escape. Corn litter (depth of litter layer approximately 5 cm, 9 g dry wt.) was placed on the surface of the soil of all mesocosms. This amount of litter corresponded to 3.42 g C and 0.16 g N per mesocosm. Litter was not replenished during the course of the experiment to mimic field conditions. All mesocosms received 1.25 cm of water at the onset of the experiment. The experiment consisted of four treatments: I. No earthworms present (CTRL); II. Artificial burrows containing no plant residues (ARTF); III. Artificial burrows containing plant residues (LEAF); and IV. Earthworms present (WORM).
Each treatment was replicated three times, with a total of 24 replicates set up initially for each treatment. CTRL mesocosms consisted of soil with corn litter on the soil surface. Artificial burrows (ARTF) were prepared using a 9-mm o.d. glass tube (containing a 4-mm diam. glass rod) that was kept in place while the column was packed. Two glass tubes per mesocosm (45 cm long) containing glass rods were placed directly across from each other and 20° off vertical, reaching to approximately 5 cm from the bottom of the core. The glass tube was removed after the column was packed and the 4-mm diam. glass rods remained inside the artificial burrow. To simulate litter in the burrows (LEAF), we wrapped a corn leaf (cut lengthwise, 1.5 g dry wt.) around the length of each of the two 4-mm diam. rods before packing the column. This corresponded to 0.98 g C and 0.046 g N per LEAF mesocosm. The leaf was secured in place with three loosely positioned nylon cable ties. The WORM treatment consisted of adding three individuals of L. terrestris L. per mesocosm, representing a population density of 370 worms m-2.
Mesocosms were incubated in the laboratory at 18 to 20°C. All mesocosms received 2 cm of water on Weeks 1, 3, 7, and 10. A total of 9.25 cm of water was added per mesocosm, corresponding to approximately one-third of the average precipitation from March to June in Kingston, RI (NOAA, 1975).
Sampling
Each treatment was sampled destructively at 0, 1, 3, 5, 7, 10, 13, and 16 wk of incubation. Litter was removed from the soil surface, weighed, placed in a sealable plastic bag, and stored in the dark at 4°C.
We defined burrow soil as that within 5 mm of a macropore wall (regardless of its origin). Midden soil was considered to be part of the burrow soil sample. Roughly equivalent amounts of burrow soil were sampled for WORM, LEAF, and ARTF treatments. Cores were sectioned (10 cm) and soil removed from the burrow walls by excavation with a spatula. Loose plant residues found in the burrows were removed using tweezers, weighed, placed in a sealable plastic bag, and stored in the dark at 4°C. Burrow and bulk soil were kept separate, with soil from different depths pooled, placed in sealable plastic bags and stored at 4°C. Earthworms were placed in a tray with water for approximately 15 min to remove soil particles from their surface. The worms were removed from the water, dried by placing on a tray lined with paper towels for approximately 15 min. and their weight recorded.
Carbon Mineralization
Soil (approximately 2 g wet wt.) was placed in a 20-mL glass serum bottle and the vial sealed with a rubber septum and an aluminum crimp collar. Carbon dioxide in the headspace of the vials was sampled and analyzed after incubation in the dark at 20°C for 24 h. Headspace gases were analyzed by gas chromatography. A 500-µL sample of headspace gases was removed by displacement using an automated headspace sampler (model 7000, Tekmar, Cinncinati, OH) and injected into a gas chromatograph (model 14A, Shimadzu, Columbia, MD). Carbon dioxide was separated using a Porapak Q column (305 cm length), reduced to CH4 by passing the gas over a heated Ni catalyst in the presence of H2, and quantified using a flame ionization detector.
Carbon Pools
Dissolved organic C was determined by extracting a 10-g soil sample with 20 mL of 0.5 M K2SO4 solution. The extract was passed through a Whatman No. 42 filter and the C concentration of the filtrate determined using a Shimadzu Total Organic Carbon Analyzer (model 5000A). Total C concentration of soil was determined using a Carlo Erba N/C analyzer. Soils in the Enfield series do not contain carbonates (Soil Survey Staff, 1981).
Nitrogen Pools
Levels of NO-3 and NH+4 in soil were determined by extracting 1 g of moist soil with 10 mL of a 2M KCl solution, followed by filtration and determination of NO-3 and NH+4 concentrations by automated colorimetric analysis using an Alpkem Flow Solution IV (Alpkem Corp., College Station, TX). Total inorganic N was defined as the sum of NO-3 and NH+4 in KCl extracts. Nitrite was not included in this pool because analyses of a subset of samples indicated that there were no detectable levels of NO-2 in soil from any of the treatments. Total N concentration of soil was determined using a Carlo Erba N/C analyzer.
Moisture Content
Moisture content of soil and plant materials was determined gravimetrically by drying to a constant weight at 105 and 65°C, respectively.
Statistical Analyses
Statistical analyses were performed using a one-way analysis of variance (P < 0.05 unless stated otherwise). Significant differences between means were determined using a pair-wise multiple comparison procedure (Student-Newmann-Keuls method). Correlations between litter remaining in WORM or LEAF treatments and C (soil C, dissolved organic C [DOC], C mineralization) and N (soil N, inorganic N, NH4–N, NO3–N) were evaluated using the Pearson product moment method.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Soil Moisture
Soil water content increased equally for all treatments during the first week of incubation (Fig. 1). Soil moisture in all treatments tended to decrease with time, the pattern differing among treatments. Moisture content was often higher in the WORM than in the CTRL treatment. Except for the LEAF treatment, for which significantly lower soil moisture levels than in CTRL treatments were observed on Weeks 3 and 7, moisture levels were lower in CTRL than in all other treatments, especially during the final 6 wk of the incubation.
Carbon Dynamics
Soil had an initial C level of 9.7 g kg-1, with small fluctuations observed in CTRL and ARTF treatments during incubation (Fig. 2) . By contrast, soil C increased in LEAF and WORM treatments on Weeks 1 and 3 (Table 1). The highest level of soil C, 13.6 g kg-1, was observed in the WORM treatment on Week 3. Thereafter the level of soil C in WORM treatments remained relatively constant and was significantly higher than in CTRL, ARTF, and LEAF treatments between Weeks 3 and 10. During this time soil C in the LEAF treatment was also generally higher than in CTRL and ARTF treatments, but lower than for WORM treatments, although differences were not statistically significant. The level of soil C in the WORM treatment declined after incubation for 10 wk, with no statistically significant differences among any of the treatments observed at 16 wk. No statistically significant correlation was observed between litter (e.g., mass of litter remaining aboveground or in the burrow) and soil C for either the WORM or LEAF treatments (Table 2).
|
|
|
The rate of C mineralization at the onset of the experiment was 6 µg CO2–C g-1 d-1 (Fig. 2). Significant differences in C mineralization were observed between CTRL and ARTF treatments only on Week 7, the rate otherwise remaining constant in both treatments (Fig. 2; Table 1). By contrast, the rate of C mineralization in the WORM treatment increased markedly on Weeks 1 and 3, declining with further incubation. The highest rate, 93 µg CO2–C g-1 d-1, was observed in the WORM treatment on Week 3. Carbon mineralization rates in the WORM treatment were significantly higher than in CTRL, ARTF, and LEAF treatments on most sampling dates. However, by Week 16 of the experiment there were no significant differences in rate among treatments. Carbon mineralization in the LEAF treatment was significantly higher than in the ARTF treatment on Weeks 7 and 10. The C mineralization rate was significantly and positively correlated to litter resources in the WORM, but not the LEAF treatment (Table 2).
Nitrogen Dynamics
Total soil N was initially 0.57 g kg-1, with values remaining relatively constant for the duration of the experiment in CTRL and ARTF treatments (Fig. 3)
. WORM and LEAF treatments had identical levels of soil N (0.82 g kg-1), which were greater than the CTRL treatment at Week 3, with levels of N in these treatments diminishing subsequently with incubation time. Soil N decreased to levels in CTRL treatment by Week 5 in the LEAF treatment and by Week 10 in WORM treatments. No significant differences in soil N were observed among treatments at 13 or 16 wk (Table 1). No statistically significant correlation was observed between litter resources and soil N in either WORM or LEAF treatment (Table 2).
|
Soil nitrate content at the onset of the experiment was 1.4 µg NO3–N g-1 (Fig. 3). Levels of nitrate increased steadily in all treatments during the first 7 wk of incubation, remaining relatively stable for the remainder of the experiment (Fig. 3). The initial rate of increase in nitrate level was highest in the WORM treatment, followed by LEAF, ARTF, and CTRL treatments. The highest levels of nitrate were observed in the WORM treatment on Week 7, when the mean value was 41 µg NO3–N g-1. Nitrate levels were significantly higher in WORM soil than in either ARTF or CTRL soil on all sampling dates and were higher than in LEAF treatments on most dates. Levels of nitrate were significantly higher in ARTF and LEAF treatments than in CTRL treatments on half of the sampling dates. There was a statistically significant, negative correlation between levels of nitrate in burrow soil and litter in both the WORM and LEAF treatments (Table 2).
Ammonium concentrations of 20 µg NH4–N g-1 were observed at the beginning of the experiment (Fig. 3). Ammonium concentration increased significantly in WORM and LEAF treatments within 1 wk of incubation relative to CTRL and ARTF treatments (Fig. 3; Table 1). The highest levels of ammonium recorded were observed for WORM soil on Week 1, when the mean value was 28 µg NH4–N g-1. Subsequently, NH+4 in the WORM and LEAF treatments declined to levels that were statistically indistinguishable from those in ARTF and CTRL treatments for the remainder of the experiment (Table 1). At the end of the 16-wk incubation period, the level of NH+4 in the CTRL treatment was 4 µg NH4–N g-1, only 20% of that observed initially. A statistically significant positive correlation was observed between levels of ammonium in burrow soil and litter resources only for the WORM treatment (Table 2).
The relative contribution of nitrate to the inorganic N pool in WORM soil increased with time, with nitrate making up between 80 and 90% of the pool as the experiment progressed. In the LEAF treatment soil nitrate constituted between 70 and 80% of the inorganic N pool. By contrast, the inorganic N pool in ARTF and CTRL treatments was composed primarily of ammonium, with the composition remaining relatively stable during the course of the experiment.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
There was a steady decline in worm weight during the course of the experiment, even though they were feeding actively, as indicated by the disappearance of litter aboveground (Fig. 1). Worm weight loss may have been a result of desiccation. The relatively low quality and rapid depletion of corn litter likely also contributed to weight loss.
Carbon Dynamics
The effects of L. terrestris on C pools and processes in burrow soil were short-lived. The difference in soil C between CTRL and WORM, and CTRL and LEAF was greatest at 3 wk, and then declined steadily (Fig. 2). The time course of soil C was probably a result of two opposing processes: (i) translocation of litter and (ii) C mineralization. In the beginning of the experiment, translocation may have been greater than C mineralization, resulting in accretion of soil C. However, toward the end of the incubation period litter became limiting in the WORM and LEAF treatments and soil C decreased. The total amount of C released as CO2 between Weeks 10 and 16 was 0.6 mg C g-1 soil, whereas the amount of C lost from the soil C pool during the same period was 3.5 mg C g-1 soil. Thus, while C mineralization may have contributed to the loss of soil C during this period, it only accounted for 1/6 of the loss. We measured C mineralization by incubating subsamples of soil taken from the mesocosms. It is possible that burrow C mineralization rates were higher in situ.
In general, the short-term nature of the effects of earthworms on soil C mirrored the findings of Devliegher and Verstraete (1995) and agreed with the suggestions of Lavelle et al. (1989), who argued that earthworms appeared to favor short and rapid, rather than long-term, turnover of organic matter and nutrients. The dynamics of burrow soil C in the WORM treatment may explain the wide range of C enrichment values observed for anecic earthworm burrow soil (e.g., Lee, 1985), since it appears that enrichment is a function of burrow age, which is seldom considered when such values are reported.
The strong positive correlation observed between litter resources and C mineralization in the WORM treatment (Table 2) indicates that mineralization was dependent on litter removal by L. terrestris. The short-term effects of LEAF treatment on C pools were generally similar in dynamics and direction to those for WORM treatment, although of lesser magnitude. This suggests that litter enrichment in burrows by itself was at least partially responsible for the effects of anecic earthworms on C dynamics in burrow soil. However, the effect of litter enrichment on C mineralization was clearly enhanced by the presence of L. terrestris. Enhanced C mineralization in the WORM treatment may be primarily associated with the decomposition of earthworm-derived mucus, which contains water-soluble, readily decomposable organic compounds (Edwards and Bohlen, 1996) but which was absent in the LEAF treatment.
The ARTF treatment did not have a significant effect on any of the C variables measured, indicating that the attributes associated with a macropore did not contribute to C dynamics in burrow soil. Kretzschmar and Monestiez (1992) found that the decomposition of labeled plant material in soil can be controlled by the presence of a burrow system when gas diffusivity in the soil limits gas exchange from the soil to the surface, with no direct effect of earthworm physiological activities observed on decomposition. It appears that C dynamics in this soil were not limited by gas diffusion.
Nitrogen Dynamics
Dynamics of NH+4 and NO-3 appeared to be coupled to litter disappearance, as indicated by temporal changes in these variables and the correlations observed between the inorganic N pools and the amount of litter remaining aboveground (Table 2). Nitrogen in worm exudates likely also contributed to inorganic N dynamics.
The persistence of NO-3 in WORM burrow soil may have resulted from a combination of biotic and abiotic factors. Plants, which constitute a major sink for soil nitrate, were not present in the mesocosms. The C/N ratio of litter (21:1) and N-enriched earthworm exudations (Edwards and Bohlen, 1996)—both of which would favor net N mineralization—tend to preclude microbial immobilization. Similarly, leaching was unlikely, given that the soil remained unsaturated (Fig. 1) during the course of the experiment. The microbial ecology of anecic earthworm burrows also favors the accretion of NO-3 in the absence of plants. Parkin and Berry (1999) observed that the drilosphere of L. terrestris was enriched in nitrate, ammonium, and soluble organic C and that it had elevated populations of nitrifying and denitrifying bacteria relative to non-drilosphere soil. However, the rate of denitrification was 1000 times lower than the rate of nitrification, suggesting that losses of nitrate from burrow soil via denitrification may be minimal (Parkin and Berry, 1999). Singer et al. (2001) suggested that the population of ammonia oxidizers in soil might be stimulated by the activities of Pheretima hawayana, an anecic earthworm.
Data from ARTF burrows suggests that the physical structure created by burrowing also contributed to the effects of anecic earthworms on nitrate levels in burrow soil. Because no resources were added intentionally to ARTF burrows, any effects of this treatment would be expected to be limited to those associated with the physical attributes of a macropore, such as preferential transport of dissolved materials and enhanced gas exchange. Significantly higher levels of NO-3 in ARTF soil than in CTRL soil suggest that the contribution of a macropore to the effects of anecic earthworm on soil inorganic N was through enhancement of the physical conditions conducive to nitrate production, such as higher levels of O2 and adequate moisture in burrow soil. Kirkham (1982) calculated that greater concentrations of O2 should be found in soil with earthworm burrows. Devliegher and Verstraete (1997) observed that O2 penetrated to a greater depth in soil with earthworms than in soil without earthworms, supporting our contention that enhanced O2 levels may be responsible for nitrate in ARTF burrow soil. Singer et al. (2001) also observed enhanced gas diffusion in soil amended with an anecic earthworm. Studying macropores of unspecified origin under field conditions, Vinther et al. (1999) found higher levels of nitrate in macropores than in bulk soil and higher levels of microbial biomass C and water soluble organic C.
Dynamics of inorganic N in LEAF burrows suggest that, unlike the WORM treatment, there was little change in this pool with incubation time. Although the litter placed in LEAF treatments had the same origin, and thus the same C/N ratio, as that in the WORM treatment, differences in litter quality may be responsible for the treatment effects. Bohlen et al. (1997) reported that the C/N ratio of corn litter in L. terrestris middens was lower than that of surrounding litter. Such selective feeding suggests that the quality of litter in burrows in the WORM treatment may well have been different from that in the LEAF treatment. Furthermore, additional inputs of N-rich substances excreted by earthworms—including ammonium, urea, and mucoproteins (Edwards and Bohlen, 1996) would have favored net N mineralization. These excretions may thus be partly responsible for the higher accumulation of inorganic N in WORM treatments relative to LEAF treatments. Differences in pathways for the loss of inorganic N, such as denitrification and diffusion, seem unlikely to explain these results, given that the conditions that would support these pathways were either identical (e.g., soil moisture) or more prevalent (e.g., greater inputs of C, higher respiration rates) in the WORM than in the LEAF treatment.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 14, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
hot spots
in soils. Soil Biol. Biochem. 33:729–738.This article has been cited by other articles:
|
|
 |
|
  M. C. Savin, J. H. Gorres, and J. A. Amador Microbial and Microfaunal Community Dynamics in Artificial and Lumbricus terrestris (L.) Burrows Soil Sci. Soc. Am. J., January 1, 2004; 68(1): 116 - 124. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
0.100.