Printer Friendly

Spatiotemporal distributions of bacterivorous nematodes and soil resources in a restored riparian wetland.

INTRODUCTION

Soils are characterized by a high degree of spatial variability, even in intensively managed ecosystems (Robertson and Freckman 1995, Robertson et al. 1997). Within large-scale patches shaped by physicochemical conditions and human activity, biotic processes create soil variability at nested smaller scales (Robertson and Gross 1994, Stark 1994). Spatial variability is not only structural, but also essentially functional in the soil ecosystem (cf. Legendre 1993). For instance, nutrient cycling "hotspots" are determined by spatial and temporal variability in soil biotic and abiotic factors (Moore and De Ruiter 1991, Beare et al. 1995). Furthermore, spatiotemporal variability at various scales is thought to be one of the principal explanations for the remarkable diversity of coexisting soil animal species, many of which exhibit only limited diet and habitat specialization (Giller 1996). Unfortunately, spatial data for soil animals are largely lacking, especially beyond large taxonomic and functional groups. Moreover, the temporal variability in spatial patterns of both soil fauna and resources is usually unknown.

We investigated the spatiotemporal variability in soil animal species distributions, and linkages to heterogeneity in soil resources, using bacterivorous nematode species as model organisms. Among soil fauna, bacterivorous nematodes are key participants in soil processes. Most importantly, they enhance nutrient cycling rates through feeding on microbes, excreting labile nutrients, and transporting microbial propagules (Hunt et al. 1987, Freckman 1988). Robertson and Freckman (1995) found that bacterivorous nematode abundance showed a high degree of spatial variability across an agroecosystem. Although the bacterivorous nematode trophic group is a common unit of analysis in soil food web studies (e.g., Moore et al. 1996), it contains many species, some of which have shown interspecific differences in metabolic rates (Anderson et al. 1981, Schiemer 1983, Ferris et al. 1995), reproductive capacity (Anderson et al. 1981, Ferris et al. 1996), colonizing ability (Bongers 1990, Ettema and Bongers 1993), and preference for temperature and moisture conditions (Sohlenius 1985). Hence, it can be expected that different bacterivorous species could exhibit partly divergent, and temporally variable, spatial distributions. Such spatiotemporal differences in population patterns would facilitate species coexistence and contribute to variability in nutrient cycling rates, at various scales in space and time.

We here report the field- and season-scale distributions of bacterivorous nematode populations in a restored riparian wetland in the coastal plain of Georgia (USA), the site of an ongoing nitrogen cycling study by Vellidis et al. (1993). We also quantify patterns of various soil properties and processes, and attempt to relate those to nematode distributions. Soil moisture, microbial respiration, inorganic and total nitrogen, and carbon were measured in the same soil samples as nematodes, in four seasons. Distributions of nematodes and soil properties were analyzed with geostatistical methods, proven to be powerful tools for detecting and quantifying spatial variability in soil ecosystems (e.g., Robertson et al. 1997).

METHODS

Study site

We conducted this study in the southeastern coastal plain, near Tifton, Georgia, USA, in a 0.7-ha restored riparian wetland drained by an intermittent first-order stream [ILLUSTRATION FOR FIGURES 1 AND 2 OMITTED]. The underlying soil is classified as an Alapaha loamy sand (loamy, siliceous, thermic Arenic Plinthic Paleaquult), and is locally flooded during winter. The wetland was deforested in 1985, served as wet pasture from 1986 to 1990, and was restored in [TABULAR DATA FOR TABLE 1 OMITTED] early 1991. Revegetation was in accordance with Riparian Forest Buffer specifications of the U.S. Department of Agriculture Forest Service for Zones I and 2 (Welsch 1991) [ILLUSTRATION FOR FIGURE 2 OMITTED]. Trees in zone 1 (bordering the stream channel) are yellow poplar (Liriodendron tulipifera L.), swamp black gum (Nyssa sylvatica var. biflora (Wait.) Sarg.), and green ash (Fraxinus pennsylvanica (Borkh.) Sarg.), whereas zone 2 consists of slash pine (Pinus elliottii Engelm.). In the southwest area no trees were planted, but a few volunteer black willow (Salix nigra) occur. The understory vegetation throughout the wetland is a mixture of native wetland grasses (mainly Paspalum sp.) and rushes (Juncus sp.). At the time of sampling, trees were 2.5-3.5 yr old. The wetland receives shallow groundwater, surface runoff, and spray drift from a center pivot manure application area used for forage production, bordering the site at the southwest end [ILLUSTRATION FOR FIGURE 1 OMITTED] (Vellidis et al. 1993). The amount of manure N available for transport into the wetland, i.e., that which is not taken up by crops or denitrified in the upland area, is estimated to be [less than]100 kg N[center dot][ha.sup.-1][center dot][yr.sup.-1] (Lowrance et al. 1995).

Soil sampling

We collected soil samples on 7 November 1993 and on 24 February, 29 April, and 25 August 1994. On each date, samples were taken within a 1-m radius of 24 spatial locations [ILLUSTRATION FOR FIGURE 2 OMITTED]. A network of shallow groundwater wells (Vellidis et al. 1993) provided the (most practical) spatial reference points for selection of sampling locations, which we established [approximately]3 m from selected wells. When possible, sampling was kept at least 2 m from tree boles. In November 1993 one location could not be sampled because of standing surface water. On the August date 12 extra points were sampled to bridge large gaps between locations [ILLUSTRATION FOR FIGURE 2 OMITTED]. Hence, the final data set contained 107 spatiotemporal observations. Spatial distance between samples ranged from 4.5 to 106.5 m.

Every sample consisted of two soil cores of 2.2-cm diameter and 15-cm length, taken within [approximately]40 cm from each other, with a corer containing a removable inner plastic tube. In the laboratory, the intact soil cores were split lengthwise. Two halves from different cores (of the same sample) were combined to form a bulk core for nematode analysis. The two remaining halves were mixed and subsampled for measurement of moisture, microbial respiration, inorganic N, and organic matter content.

Biological and chemical measurements

We extracted nematodes from the bulk core, which included soil, roots, and litter, using Cobb's decanting and sieving method combined with a blender-cotton-wool filter procedure (Southey 1986, Schouten and Arp 1991). After fixation in 4% formaldehyde nematodes were counted and identified using an inverted microscope. We performed the spatiotemporal analysis for eight bacterivores (Table 1), one of which was identified at the superfamily (Rhabditinae), and seven at the genus level (species-level identification was virtually impossible as site-specific taxonomic keys were lacking and samples were dominated by juveniles). Since our sampling plan was not designed to evaluate small-scale variability ([less than]4.5 m), we converted counts to individuals/[m.sup.2] (15 cm depth), to analyze data at a standard ecosystem level (cf. Robertson and Freckman 1995). This conversion does not affect the shape (relative structure and range) of the spatial semivariogram nor the spatial pattern predicted by kriging, although it raises kriging variance.

We measured microbial respiration in 10-g root-free subsamples, after incubation at 22 [degrees] C for 2 h, in a continuous airflow system (Cheng and Virginia 1993). Evolving C[O.sub.2] was measured on an infrared gas analyzer (LI-COR 6252). Results (as mass of C[O.sub.2]-C released per unit soil dry mass per hour) were converted to daily rates on an areal basis (mass of C[O.sub.2]-C released per unit area per day, to 15 cm depth) using an average soil bulk density of 1.28 g/[cm.sup.3] (R. Lowrance, unpublished data).

We determined soil total carbon (C) and nitrogen (N) by Dumas combustion on a Carlo-Erba C/N Analyzer (Page et al. 1982). Data units are gravimetric percentages. Soil nitrate N and ammonium N were measured colorimetrically, in 50 mL of 2-mol/L KCl extracts of 10-g subsamples, using an Alpkem Autoanalyzer (Page et al. 1982). Results (as mass of N per unit soil dry mass) were converted to areal units (mass of N per unit area, to 15 cm depth) and log-transformed before analysis.

Statistical analysis

We performed standard parametric analyses and geostatistical analyses in SAS, using the correlation (CORR), regression (REG), and interactive matrix (IML) procedures (Statistical Analysis Systems Institute 1988). See the Appendix for information on obtaining supplementary material.

Semivariogram estimation. - We estimated spatial semivariograms by pooling over the sampling dates (Cressie 1993:274). This procedure, which yields more precise inferences than estimation based on a single sampling event, treats observations from different dates as replicate realizations of the spatial process, assuming that the spatial correlation structure does not change over time. We explored the possibility of including temporal correlation in the semivariogram model, since kriging potentially could be improved by basing predictions on data from temporal as well as spatial neighbors. However, we were unable to detect temporal correlation in the data, as most temporal semivariograms (semivariances between observations at different dates, at the same spatial locations) showed pure nugget effects (temporally independent variance). We suspect that sampling in contrasting seasons limited the similarity between close temporal neighbors, and that nematode populations were primarily correlated at time scales shorter than sampled. In addition, it appears that four dates (which provide only six temporal distance classes) are hardly sufficient to detect and exploit temporal correlation.

We estimated the pooled semivariance [Gamma](h), for each of nine spatial distance intervals h, using the nonparametric estimator

[Mathematical Expression Omitted] (1)

where z([s.sub.it]) is the measured sample value at spatial location [s.sub.i], at sampling date t (t = 1, . . ., 4); [s.sub.i] and [s.sub.j] are separated by average distance h; [n.sub.h], is the number of pairs ([s.sub.i], [s.sub.j]) in distance class h at sampling date t; and [N.sub.h] is the total number of pairs ([s.sub.i], [s.sub.j]) in distance class h summed over four sampling dates. Thus, [Mathematical Expression Omitted] estimates the average semivariance between points separated by distance h, averaged over four dates. The [N.sub.h] values for the first three distance classes, centered at 5, 15, and 25 m, were 33, 200, and 250 pairs, respectively.

Most variates, except soil moisture and Monhystrella sp., exhibited large-scale spatial trends across the wetland. For such nonstationary data, semivariograms fail to level off with increasing intersample distance. However, for optimal and unbiased prediction (kriging) in the local neighborhood the large-scale trends have to be removed. This was achieved by replacing the original data z([s.sub.it]) in Eq. 1 by the residuals [Mathematical Expression Omitted] from ordinary least squares (OLS) regression of z([s.sub.it]) against x, y coordinates of the sample sites &, (cf. Cressie 1993: 151, 165-170). The models fitted to these residuals-based semivariograms were used in universal (rather than ordinary) kriging routines.

Semivariogram modeling. - We fitted the estimated semivariograms by weighted least squares (Cressie 1993:98-100) to mostly exponential functions, which generally had smaller residual sums of squares and better fit at small distances than other models. Gaussian functions fitted better to semivariograms of nitrate N and ammonium N. Most models were fit across a range of 0-85 m, i.e., 80% of the maximum separation distance. However, models for moisture, microbial respiration, Chronogaster sp., and Rhabdolaimus sp. were fit across a smaller range (0-65 m) due to erratic semivariances at greater distance intervals.

Kriging. - For all variates showing spatial correlation and large-scale trends, we predicted unsampled points by universal kriging (Cressie 1993:151-157), using the models fitted to residuals-based semivariograms. The exception was moisture, which was predicted using ordinary kriging as its data contained no large-scale spatial trends. Unlike ordinary kriging, universal kriging includes known functions of (fixed) explanatory variables, in our case x, y coordinates. It is the optimal routine for kriging nonstationary data. Isopleths were drawn in DeltaGraph 4.0 for Windows (DeltaPoint 1996).

Tests for spatial trends. - To further evaluate the large-scale spatial trends in the data, we performed two-sided t tests for generalized least squares (GLS) estimates of the partial slopes of x, y coordinates. Although OLS regression gives unbiased estimates of partial slopes, their standard errors, as estimated by common regression software, are biased in the presence of spatial correlation. GLS estimates of regression coefficients use information from the spatial correlation matrix (which, in our case, is derived from the model fitted to the residuals-based semivariogram). By giving more weight to isolated sites than to sites in crowded locations, which present redundant information, GLS regression coefficient estimates are more precise than OLS estimates, and estimates of their standard errors are unbiased (cf. Cressie 1993:20-24). GLS estimates can be used to test effects of independent variables, such as x, y coordinates.

Correlations. - With spatially dependent data, Pearson's correlation coefficients are valid, but their P values are incorrect because variance estimation is biased (Legendre 1993). In this paper we provide Pearson's [TABULAR DATA FOR TABLE 2 OMITTED] correlation coefficients as quantitative indications of correlations between nematodes and soil resources, without granting their statistical significance.

RESULTS

We extracted an abundant and diverse nematode assemblage from the wetland soil. Of the 68 taxa identified, 25 were classified as bacterial feeders (C. Ettema, unpublished data). We here report an analysis of the eight most dominant bacterivores, which together represented 85% of total bacterivores on average (Table 1). Abundances were sizably higher than found in the scant literature on freshwater wetland nematode fauna (Overgaard Nielsen 1949, Cox and Smart 1994; Table 1). We estimated maximum local densities of [greater than]1 x [10.sup.6] individuals/[m.sup.2] for the most dominant taxon, Acrobeloides. Eumonhystera was least numerous, averaging 3.6 x [10.sup.4] individuals/[m.sup.2]. Population sizes were highly variable across the site and seasons, with overall coefficients of variation (CVS) ranging from 80% for Prismatolaimus (the only taxon present at all locations, at all dates) to 153% in Rhabditinae. Maximum local and average abundances were observed at different seasons for different taxa.

[TABULAR DATA FOR TABLE 3 OMITTED]

Microbial and nutrient pools were generally less variable, with overall cvs of 30-36%, except inorganic N, which had cvs up to 141% (Table 2). Wetland soil moisture varied from dry to fully saturated conditions (6-56%), Microbial respiration averaged 9.8 g C[O.sub.2]-C[center dot][m.sup.-2][center dot][d.sup.-1], and varied over a range similar to that reported for a poorly drained riparian forest soil (Groffman et al. 1992). Nitrate N and ammonium N levels were on average [less than] 1 g/[m.sup.2]. Soil total C and N ranged from 0.58% and 0.02% of soil dry mass, to 2.93% and 0.14%, respectively. Mean soil organic matter C/N ratio was 18.9.

Semivariogram analysis

Pooled semivariogram analysis showed that a substantial portion of the variation in nematode populations and soil resources was spatially dependent. For nematodes, average sample population variances were 67-99% spatially structured, except for Monhystrella and Heterocephalobus, which showed no spatial dependence at the scale examined (Table 3, [ILLUSTRATION FOR FIGURE 3 OMITTED]). The distance over which nematode populations were spatially dependent varied from 15 m, for Chronogaster, to 67 m for Acrobeloides.

For microbial and nutrient pools, 36-99% of average sample population variance was spatially dependent (Table 4, [ILLUSTRATION FOR FIGURE 4 OMITTED]). Most soil resources were autocorrelated within 10-20 m, except nitrate- and ammonium N, which varied over larger distances (56-84 m).

Spatial patterns in November 1993

Isopleths drawn for November 1993 show that nematode spatial patterns were overlapping between some taxa, while being notably dissimilar among other taxa. Acrobeloides, Rhabditinae, and Eumonhystera were most abundant in the northwest corner of the wetland [ILLUSTRATION FOR FIGURE 5A-C OMITTED], whereas Chronogaster populations concentrated in a hotspot in the south, Prismatolaimus was most abundant in the west, and Rhabdolaimus was most numerous in the east [ILLUSTRATION FOR FIGURE 5D-F OMITTED]. Most of these trends were statistically significant (Table 5).

Unlike the divergent nematode spatial trends, soil resource patterns were mostly overlapping ([ILLUSTRATION FOR FIGURE 6 OMITTED], Table 6). Microbial respiration, soil C and N, nitrate N, [TABULAR DATA FOR TABLE 4 OMITTED] and ammonium N all were higher in the northern part of the wetland. In addition, small-scale patterns of soil C and N concentration and microbial respiration were remarkably similar [ILLUSTRATION FOR FIGURE 6A-C OMITTED]. Moisture was highest in the south central area, around the intermittent stream and in seepage areas [ILLUSTRATION FOR FIGURE 6D OMITTED].

Temporal variation in spatial patterns

Most large-scale spatial trends in nematode patterns observed in November 1993, persisted at later sampling dates. Acrobeloides, Rhabditinae, and Eumonhystera remained significantly more abundant in the north and northwest portions of the wetland throughout the four seasons (except in August 1994 for Acrobeloides) (Table 5). Prismatolaimus appeared more abundant in the west on all dates, yet this value was only statistically significant in February 1994. Rhabdolaimus maintained more numerous populations in the east, though this trend ceased to be significant at the last sampling date. Within these relatively constant large-scale spatial trends, small-scale patterning was notably dynamic, as illustrated by the seasonal isopleth series for Acrobeloides and Chronogaster [ILLUSTRATION FOR FIGURES 7 AND 8 OMITTED].

Spatial patterns for soil resources were generally more static, which was also observed for two successive seasons by Goovaerts and Chiang (1993) for similar variates. Expectedly, soil total C and N patterns were extremely stable over time (Table 6). Though values typically varied across dates, soil moisture distribution retained its general pattern, with highest levels around the intermittent stream and in seepage areas (cf. [ILLUSTRATION FOR FIGURE 6D OMITTED]). Inorganic N patterns were consistent with November observations (cf. [ILLUSTRATION FOR FIGURE 6E, F OMITTED]), though absolute values were lower in February and April, compared to November and August dates (Table 2). Microbial respiration was generally higher in February and April, and locations of hotspots were dynamic [ILLUSTRATION FOR FIGURE 9 OMITTED].

Relations between nematode and resource patterns

To investigate relations between bacterivorous nematode distributions and soil resource patterns, we visually compared isopleths and calculated correlation coefficients (Table 7). In November 1993, peak abundances of Acrobeloides, Rhabditinae, and Eumonhystera appeared to occur where nitrate N was highest [ILLUSTRATION FOR FIGURES 5A-C, 6E OMITTED]. Correlation coefficients were 0.66, 0.79, and 0.47, respectively (Table 7). Prismatolaimus was negatively correlated with soil moisture (r = -0.57), its isopleth resembling the "negative" of the moisture distribution [ILLUSTRATION FOR FIGURES 5E, 6D OMITTED]. Chronogaster, in contrast, was most abundant in a wet area in the south (r = 0.40). Interestingly, no strong correlations were found for any of the bacterivorous nematodes with microbial respiration. Comparing nematode isopleths with the vegetation map [ILLUSTRATION FOR FIGURE 1 OMITTED], it appears that taxa were not limited to specific vegetation types.

Between sampling dates, correlations varied considerably [TABULAR DATA FOR TABLE 5 OMITTED] [TABULAR DATA FOR TABLE 6 OMITTED] (Table 7). The positive correlation between nitrate N and Acrobeloides and Eumonhystera found for November 1993 weakened in course of the study, while Rhabditinae kept a consistently positive association with this nutrient. Overall, associations between nematode and soil resource patterns seemed weak, as we found only seven of 144 coefficients to be [greater than or equal to]0.50.

DISCUSSION

Our data demonstrate that spatial aggregation is an important feature of bacterivorous nematode populations and soil resources in the wetland examined. These results corroborate findings of other studies, which have found spatial dependence for a wide array of soil biotic and abiotic properties over comparable spatial ranges (e.g., Poier and Richter 1992, Fromm et al. 1993, Smith et al. 1994, Schlesinger et al. 1996, Robertson et al. 1997). Whereas most of these studies provide a spatial "snapshot" of one moment in time, our results from replicated spatial sampling show that temporal variation adds a significant dimension to spatial distributions, particularly for soil biological populations [ILLUSTRATION FOR FIGURES 7-9 OMITTED].

Building upon Robertson and Freckman's (1995) spatial analysis of total bacterivorous nematode abundance, our study demonstrates that spatial aggregation occurs in populations of individual bacterivorous taxa, and that spatial patterns are different among them (Table 3, [ILLUSTRATION FOR FIGURE 5 OMITTED]). For six of eight bacterivores analyzed, spatial autocorrelation over a range of 15-67 m accounted for a major proportion (67-99%) of total sample variance. The two remaining taxa, Monhystrella and Heterocephalobus, had no spatially dependent variance, their semivariograms showing pure nugget effects [ILLUSTRATION FOR FIGURE 3E, F OMITTED]. We have no ready statistical or biological answer as to why nugget variance (which includes random sampling error and spatial variation at scales smaller than sampled) [TABULAR DATA FOR TABLE 7 OMITTED] was larger for Heterocephalobus and Monhystrella than for other taxa, except that identification error may have been greater for Heterocephalobus. Diagnostic characteristics for this genus are highly unsatisfactory (De Ley et al. 1993), and the Heterocephalobus counts possibly included several taxa that obscured each other's distributions. Unfortunately, our sampling design precludes enumeration and comparison of small-scale variability in the nematode populations analyzed. Significant small-scale aggregation (over 20-120 cm), linked with rooting patterns in sugarcane rows, has been observed for some populations of nematodes that parasitize plants (Rossi et al. 1996). At even smaller scales, meiobenthic nematodes have been found aggregated in pockets of organic debris buried in sediment ripple troughs spaced [less than]8 cm apart (Hogue and Miller 1981). We expect that similar small-scale population hotspots, associated with roots and organic particles, occur nested within the large-scale patterns observed in the wetland.

Yet, spatial aggregation at the scale assessed in our study has important implications for understanding "regional" (field-scale) nematode diversity and population dynamics. We observed, for all seasons sampled, that coexisting bacterivorous taxa exhibited divergent spatial distributions, with populations aggregating into different hotspots in the wetland ([ILLUSTRATION FOR FIGURE 5 OMITTED], Table 5). The most dominant taxa, Acrobeloides and Prismatolaimus, had more continuous spatial populations (i.e., larger ranges; Table 3) than the less abundant bacterivores, possibly reflecting the importance of regional population size for resistance against local extinction (Wright and Coleman 1993). The temporal sequence of Chronogaster isopleths provides a striking illustration of the rise and fall of local populations, while regional population size remains relatively stable (Table 1, [ILLUSTRATION FOR FIGURE 8 OMITTED]).

Assuming that field-scale patterns in nematode populations are primarily a response to patchiness and gradients in their abiotic environment (more so than a direct reflection of small-scale biotic processes such as reproduction and competition), we found surprisingly few linkages between nematode and soil resource distributions. Combining the correlation analysis with a visual comparison of the isopleths yielded a few patterns.

First, peak populations of Acrobeloides, Rhabditinae, and Eumonhystera in November 1993 occurred where nitrate N was highest [ILLUSTRATION FOR FIGURES 5A-C, 6E OMITTED], and Rhabditinae maintained a strong association with this nutrient at all dates (Table 7). Nitrate N may have been elevated because high abundances of fast growing nematodes excreted ammonium (e.g., Bouwman et al. 1994), which was then nitrified. However, nitrate levels may have also been the cause, instead of the consequence, of population growth for Acrobeloides, Rhabditinae, and Eumonhystera. Experiments in forest (Baath et al. 1978, Sohlenius and Wasilewska 1984), agricultural (Sohlenius and Bostrom 1986), and rhizosphere soil (Griffiths et al. 1992) have shown population rises of the same bacterivorous taxa in response to inorganic N addition. The mechanism thought to be responsible is that these taxa, which all have short life cycles and high fecundity, are able to quickly exploit N-limited microbial production. Although the soil of our wetland site has low standing stocks of inorganic N ([less than] 1 g/[m.2.sup]), it receives significant influx of N from the upland manure application site, most of which is rapidly denitrified (the average rate is 68 kg [N.sup.2]O-N.[ha.sup.-1].[yr.sup.-1] [Lowrance et al. 1995]). Accumulation of nitrate N in the northwest corner,of the wetland may have occurred because denitrification was limited by lower soil moisture in this area, due to local topography and draining influence of the pond [ILLUSTRATION FOR FIGURE 6D OMITTED] (Lowrance et al. 1995). To further investigate nematode-N associations and to assess which bacterivorous nematode taxa can be considered indicators of N enrichment of riparian soils, we are currently performing experimental N additions in an uncontaminated riparian forest near the site.

Second, soil moisture distributions offered a partial explanation for nematode patterns. With the exception of Chronogaster, all nematodes were least abundant in the wettest soil around the intermittent stream and in seepage areas, at all sampling dates ([ILLUSTRATION FOR FIGURES 5, 6D OMITTED], Tables 5, 7). Though nematodes can survive low-oxygen conditions in saturated soils (Poinar 1983) [ILLUSTRATION FOR FIGURE 5 OMITTED], accumulating products of anaerobic metabolism may inhibit population growth for many taxa. Chronogaster, the only taxon thriving in the wettest areas (at least in November and February), has an extremely high body surface-to-volume ratio, which facilitates oxygen uptake. A species of the genus has been found in such extreme habitats as [H.sub.2]S-rich thermomineral water in a Romanian cave (Poinar and Sarbu 1994).

Nematode spatial patterns were mostly uncorrelated with microbial respiration, soil organic matter, and vegetation type (Table 7). Lack of significant correlation between bacterivorous nematode abundance and measures of microbial biomass or respiration has been reported often (e.g., Clarholm et al. 1981, Wardle et al. 1995) and has been explained by the nonlinear nature of the trophic relationship, which includes indirect effects and positive feedbacks (Bengtsson et al. 1996). Microbial respiration was more clearly correlated with soil organic carbon (Pearson's correlations: r = 0.55, 0.50, 0.75, and 0.48 for the respective sample dates). This pattern in turn reflects an accumulation of soil organic matter downslope.

In summary, we could only infrequently explain the diverging spatial distributions of individual bacterivorous taxa as differential responses to soil resource patterns. In field surveys, correlations between nematode taxa and large-scale soil resources are never found to be particularly strong, but our values were weaker than reported in other studies (e.g., Wallace et al. 1993, Robertson and Freckman 1995), despite our care to measure all variates in the same soil sample. The young age of the wetland and vegetation (restoration actions were completed 2.5 yr before the start of our study) is one explanation for the lack of correspondence between nematode and soil resource patterns. Wardle et al. (1995) found that significant nematode-soil resource correlations emerged only 2 yr after disturbing an agricultural soil by mulching, cultivation, and herbicides. In the wetland, which has gone through more radical disturbance, nematode patterns and soil properties could take considerably longer to converge. Furthermore, temporal analysis showed that wetland nematode population patterns were considerably more dynamic over time than most soil resource distributions, which explains why the majority of correlation coefficients varied "inconsistently" between sampling dates (Table 7). Clearly, explanatory power of simple resource relationships is limited as nematode populations result from complex biotic and abiotic interactions in space and time. We realize that there are even fewer tools available to explain the observed spatial patterns in terms of small-scale population processes, because of the paucity of knowledge of basic life history characteristics for all taxa in this study except perhaps Acrobeloides and Rhabditinae (Ferris et al. 1995, 1996). Still slighter is our knowledge on the nature of interspecific interactions under field conditions (Sohlenius 1985).

Although nematode patterns remain inadequately explained, we suggest that the observed spatiotemporal divergence in bacterivorous taxa populations is highly relevant to the understanding of soil ecosystem and community processes, notably the spatiotemporal patterns of nematode-influenced nitrogen cycling rates and the maintenance of field-scale nematode diversity. We anticipate that spatiotemporal distributions of other soil fauna could exhibit similar levels of variability (albeit at slightly different scales), which would further contribute to the spatiotemporal mosaic of soil processes and the maintenance of soil biodiversity.

ACKNOWLEDGMENTS

We thank Rodney Hill, Lee West, Dorota Porazinska, and Luc Boerboom for providing help with soil sampling, Paul Hendrix for letting us use his LI-COR IR Gas Analyzer, Keith Kisselle and Bob Potter for assisting with soil analysis on the Carlo-Erba, and Liam Heneghan and two anonymous reviewers for their helpful comments on the manuscript. This work was supported by US-EPA Project Nos. 2CNW63N00D and 2CNW63M016, State and Hatch funds allocated to the Georgia Agricultural Experiment Station, Tifton, Georgia, and federal funds allocated to the USDA-ARS Southeast Watershed Research Laboratory, Tifton, Georgia.

LITERATURE CITED

Anderson, R. V., D.C. Coleman, C. V. Cole, and E. T. Elliott. 1981. Effect of the nematodes Acrobeloides sp. and Mesodiplogaster lheritieri on substrate utilization and nitrogen and phosphorus mineralization in soil. Ecology 62:549-555.

Baath, E., U. Lohm, B. Lundgren, T Rosswall, B. Soderstrom, B. Sohlenius, and A. Wiren. 1978. The effect of carbon and nitrogen supply on the development of soil organism populations and pine seedlings: a microcosm experiment. Oikos 31:153-163.

Beare, M. H., D.C. Coleman, D. A. Crossley, Jr., P. F. Hendrix, and E. P. Odum. 1995. A hierarchical approach to evaluating the significance of soil biodiversity to biogeochemical cycling. Plant and Soil 170:5-22.

Bengtsson, J., H. Setala, and D. W. Zheng. 1996. Food webs and nutrient cycling in soils: interactions and positive feedbacks. Pages 30-38 in G. A. Polis and K. O. Winemiller, editors. Food webs - integration of patterns and dynamics. Chapman and Hall, New York, New York, USA.

Bongers, T 1990. The Maturity Index: an ecological measure of environmental disturbance based on nematode species composition. Oecologia 83:14-19.

Bouwman, L. A., J. Bloem, E H. J. F. Van den Boogert, F. Bremer, G. H. J. Hoenderboom, and P. C. De Ruiter. 1994. Short-term and long-term effects of bacterivorous nematodes and nematophagous fungi on carbon and nitrogen mineralization in microcosms. Biology and Fertility of Soils 17:249-256.

Cheng, W., and R. A. Virginia. 1993. Measurement of microbial biomass in arctic tundra soils using fumigation-extraction and substrate-induced respiration procedures. Soil Biology and Biochemistry 25:135-142.

Clarholm, M., B. Popovic, T Rosswall, B. Soderstrom, B. Sohlenius, H. Staff, and A. Wiren. 1981. Biological aspects of nitrogen mineralization in humus from a pine forest podsol incubated under different temperature and moisture conditions. Oikos 37:137-145.

Cox, R. J., and G. C. Smart. 1994. Nematodes associated with plants from naturally acidic wetland soils. Journal of Nematology 26:535-537.

Cressie, N. A. C. 1993. Statistics for spatial data. John Wiley and Sons, New York, New York, USA.

De Ley, P., M. R. Siddiqi, and S. Bostrom. 1993. A revision of the genus Pseudacrobeles Steiner, 1938 (Nematoda: Cephalobidae). Part 1. Subgenus Pseudacrobeles grad. n. Fundamental and Applied Nematology 16:219-238.

DeltaPoint. 1996. DeltaGraph user guide version 4.0. DeltaPoint, Monterey, California, USA.

Ettema, C. H., and T Bongers. 1993. Characterization of nematode colonization and succession in disturbed soil using the Maturity Index. Biology and Fertility of Soils 16: 79-85.

Ferris, H., M. Eyre, R. C. Venette, and S.S. Lau. 1996. Population energetics of bacterial-feeding nematodes: stage-specific development and fecundity rates. Soil Biology and Biochemistry 28:271-280.

Ferris, H., S.S. Lau, and R. Venette. 1995. Population energetics of bacterial-feeding nematodes: respiration and metabolic rates based on [CO.sub.2], production. Soil Biology and Biochemistry 27:319-330.

Freckman, D. W. 1988. Bacterivorous nematodes and organic matter decomposition. Agriculture, Ecosystems and Environment 24:195-217.

Fromm, H., K. Winter, J. Filser, R. Hantschel, and F. Beese. 1993. The influence of soil type and cultivation system on the spatial distributions of the soil fauna and microorganisms and their interactions. Geoderma 60:109-118.

Giller, P. 1996. The diversity of soil communities, the "poor man's tropical rainforest." Biodiversity and Conservation 5:135-168.

Goovaerts, P., and C. N. Chiang. 1993. Temporal persistence of spatial patterns for mineralizable nitrogen and selected soil properties. Soil Science Society of America Journal 57:372-381.

Griffiths, B. S., R. Welschen, J. J. C. M. van Arendonk, and H. Lambers. 1992. The effect of nitrate-nitrogen supply on bacteria and bacterial-feeding fauna in the rhizosphere of different grass species. Oecologia 91:253-259.

Groffman, P.M., A. J. Gold, and R. C. Simmons. 1992. Nitrate dynamics in riparian forests: microbial studies. Journal of Environmental Quality 21:666-671.

Hogue, E. W., and C. B. Miller. 1981. Effects of sediment microtopography on small-scale spatial distributions of meiobenthic nematodes. Journal of Experimental Marine Biology and Ecology 53:181-191.

Hunt, H. W., D.C. Coleman, E. R. Ingham, R. E. Ingham, E. T. Elliott, J. C. Moore, S. L. Rose, C. P. P. Reid, and C. R. Morley. 1987. The detrital food web in a short grass prairie. Biology and Fertility of Soils 3:57-68.

Legendre, P. 1993. Spatial autocorrelation: trouble or new paradigm? Ecology 74:1659-1673.

Lowrance, R., G. Vellidis, and R. K. Hubbard. 1995. Denitrification in a restored riparian forest wetland. Journal of Environmental Quality 24:808-815.

Moore, J. C., and P. C. De Ruiter. 1991. Temporal and spatial heterogeneity of trophic interactions within below-ground foodwebs. Agriculture, Ecosystems and Environment 34: 371-397.

Moore, J. C., P. C. De Ruiter, H. W. Hunt, D.C. Coleman, and D. W. Freckman. 1996. Microcosms and soil ecology: critical linkages between field studies and modeling food webs. Ecology 77:694-705.

Overgaard Nielsen, C. 1949. Studies on the soil microfauna II. The soil inhabiting nematodes. Natura Jutlandica 2:1-131.

Page, A. L., R. H. Miller, and D. R. Keeney. 1982. Methods of soil analysis. Part 2. Chemical and microbiological properties. American Society of Agronomy, Soil Science Society of America, Madison, Wisconsin, USA.

Poier, K. R., and J. Richter. 1992. Spatial distribution of earthworms and soil properties in an arable loess soil. Soil Biology and Biochemistry 24:1601-1608.

Poinar, G. O., Jr. 1983. The natural history of nematodes. Prentice-Hall, Englewood Cliffs, New Jersey, USA.

Poinar. G. O., Jr., and S. M. Sarbu. 1994. Chronogaster troglodytes sp. n. (Nemata: Chronogasteridae) from Movile Cave, with a review of cavernicolous nematodes. Fundamental and Applied Nematology 17:231-237.

Robertson, G. E, and D. W. Freckman. 1995. The spatial distribution of nematode trophic groups across a cultivated ecosystem. Ecology 76:1425-1432.

Robertson, G. E, and K. L. Gross. 1994. Assessing the heterogeneity of belowground resources: quantifying pattern and scale. Pages 237-253 in M. M. Caldwell and R. W. Pearcy, editors. Exploitation of environmental heterogeneity by plants. Academic Press, San Diego, California, USA.

Robertson, G. P., K. M. Klingensmith, M. J. Klug, E. A. Paul, J. R. Crum, and B. G. Ellis. 1997. Soil resources, microbial activity, and primary production across an agricultural ecosystem. Ecological Applications 7:158-170.

Rossi, J.-P., L. Delaville, and P. Queneherve. 1996. Microspatial structure of a plant-parasitic nematode community in a sugarcane field in Martinique. Applied Soil Ecology 3:17-26.

Schiemer, E 1983. Comparative aspects of food dependence and energetics of freeliving nematodes. Oikos 41:32-42.

Schlesinger, W. H., J. A. Raikes, A. E. Hartley, and A. F. Cross. 1996. On the spatial pattern of soil nutrients in desert ecosystems. Ecology 77:364-374.

Schouten, A. J., and K. K. M. Arp. 1991. A comparative study on the efficiency of extraction methods for nematodes from different forest litters. Pedobiologia 35:393-400.

Smith, J. L., J. J. Halvorson, and H. Bolton Jr. 1994. Spatial relationships of soil microbial biomass and C and N mineralization in a semi-arid shrub-steppe ecosystem. Soil Biology and Biochemistry 26:1151-1159.

Sohlenius, B. 1985. Influence of climatic conditions on nematode coexistence: a laboratory experiment with a coniferous forest soil. Oikos 44:430-438.

Sohlenius, B., and S. Bostrom. 1986. Short-term dynamics of nematode communities in arable soil - influence of nitrogen fertilization in barley crops. Pedobiologia 29:183-191.

Sohlenius, B., and L. Wasilewska. 1984. Influence of irrigation and fertilization on the nematode community in a Swedish pine forest soil. Journal of Applied Ecology 21: 327-342.

Southey, J. E, editor. 1986. Laboratory methods for work with plant and soil nematodes. Minister of Agriculture, Fisheries and Food, Reference Book 402, London, UK.

Stark, J. M. 1994. Causes of soil nutrient heterogeneity at different scales. Pages 255-284 in M. M. Caldwell and R. W. Pearcy, editors. Exploitation of environmental heterogeneity by plants. Academic Press, San Diego, California, USA.

Statistical Analysis Systems Institute. 1988. SAS/STAT user's guide, Release 6.03 Edition. SAS Institute, Cary, North Carolina, USA.

Vellidis, G., R. Lowrance, M. C. Smith, and R. K. Hubbard. 1993. Methods to assess the water quality impact of a restored riparian wetland. Journal of Soil and Water Conservation 48:223-230.

Wallace, M. K., R. H. Rust, D. M. Hawkins, and D. H. MacDonald. 1993. Correlation of edaphic factors with plant-parasitic nematode population densities in a forage field. Journal of Nematology 25:642-653.

Wardle, D. A., G. W. Yeates, R. N. Watson, and K. S. Nicholson. 1995. The detritus foodweb and the diversity of soil fauna as indicators of disturbance regimes in agroecosystems. Plant and Soil 170:35-43.

Webster, R. 1985. Quantitative spatial analysis of soil in the field. Advances in Soil Science 3:2-70.

Welsch, D.W. 1991. Riparian forest buffers. Publication NA-PR-07-91. U.S. Department of Agriculture Forest Service, Washington, D.C., USA.

Wright, D. H., and D.C. Coleman. 1993. Patterns of survival and extinction of nematodes in isolated soil. Oikos 67:563-572.

APPENDIX

Supplementary material relating to the statistical analysis is available in ESA's Electronic Data Archive: Ecological Archives E079-001.
COPYRIGHT 1998 Ecological Society of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1998 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Ettema, Christen H.; Coleman, David C.; Velidis, George; Lowrance, Richard; Rathbun, Stephen L.
Publication:Ecology
Date:Dec 1, 1998
Words:6169
Previous Article:Regulation of fine root dynamics by mammalian browsers in early successional Alaskan Taiga forests.
Next Article:Nutrient acquisition from soils with patchy nutrient distributions as assessed with simulation models.
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters