Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A comparison of the spatial variability of denitrification and related soil properties in restored and natural depression wetlands in Indiana, USA

A comparison of the spatial variability of denitrification and related soil properties in... International Journal of Biodiversity Science, Ecosystem Services & Management, 2015 Vol. 11, No. 1, 36–45, http://dx.doi.org/10.1080/21513732.2014.950981 A comparison of the spatial variability of denitrification and related soil properties in restored and natural depression wetlands in Indiana, USA a b a John M. Marton *, Rinku Roy Chowdhury and Christopher B. Craft School of Public and Environmental Affairs, Indiana University, 702 North Walnut Grove Ave, Bloomington, IN 47405, USA; Department of Geography, Indiana University, 701 East Kirkwood Ave, Bloomington, IN 47405, USA We characterized and compared the variability and spatial patterns of denitrification and soil properties in two natural and two restored depressional wetlands in northern Indiana, USA. Soil properties included soil moisture content, bulk density, plant-available N (NO -N, NH -N), soil organic carbon (C), total nitrogen (N), and C:N ratio. Restored wetlands had greater 3 4 denitrification and higher spatial variability than natural wetlands. Further, restored wetlands had greater bulk density and C: N ratios. In contrast, natural wetlands had greater soil moisture, plant-available N, organic C, and total N. Similar to denitrification, soil moisture, bulk density, NO -N, total N, and C:N ratios had greater variance in restored wetlands than in natural wetlands. Denitrification and several soil properties exhibited positive global spatial autocorrelation, though trends differed between the individual wetlands for soil properties. Variogram analysis suggested little spatial structure in variables at the chosen observational scale. Denitrification hot spots were detected in natural and restored wetlands, though these hot spots did not correspond to hot spots of any of the other soil variables. Overall, spatial patterns of denitrification and soil properties differed between natural and restored wetlands and should be considered when assessing effectiveness of restored wetlands at providing ecosystem services, such as N removal and C storage. Keywords: wetland restoration; denitrification; spatial variability; soil organic carbon; spatial autocorrelation Introduction and soil texture (Mitsch et al. 2001; Bruland et al. 2003, 2006; Ullah & Faulkner 2006). Wetlands exist at the interface of upland and aquatic The spatial structure of soil properties and processes systems and provide many ecosystem services such as adds an additional level of complexity to the comparison carbon (C) sequestration, nutrient (nitrogen [N] and phos- of natural and restored wetlands. Spatial heterogeneity can phorus [P]) retention and transformation, water storage, influence above- and belowground biodiversity, which in and biodiversity support (Mitsch et al. 2001; Zedler turn can affect biogeochemical cycling of C and N 2003; Fennessy & Craft 2011). Wetlands in agricultural (Schlesinger et al. 1996; Ettema et al. 1998; Ettema & landscapes are uniquely positioned to intercept nutrients Wardle 2002). Wetland functions also can exhibit strong from agricultural fields, limiting the potential for down- spatial patterns over multiple scales, ranging from centi- stream eutrophication. In particular, wetlands are effective meters to kilometers (Ettema et al. 1998; King et al. 2004). at removing excess NO -N from surface and groundwater Determining the spatial variation in the nutrient status and draining agricultural fields (Woltemade 2000; Mitsch et al. biogeochemical processes in wetlands will allow managers 2001; Zedler 2003; Kovacic et al. 2006). to determine whether certain locations of a wetland are Wetland restoration aims to reintroduce ecosystem ser- more likely to be a nutrient source or sink (Grunwald et al. vices to the landscape. This is particularly true in the 2004, 2007). Hydrologic flowpaths into and along the agricultural landscapes of Midwestern United States wetlands influence nutrient, particularly NO -N, delivery where extensive wetland loss has occurred due to the to macrophytes and above- and belowground production, expansion of row-crop agriculture (Mitsch et al. 2001; which in turn affects distribution of soil organic matter and Zedler 2003; Fennessy & Craft 2011). Water quality microbial communities (Anderson et al. 2005). Further, improvement through N removal is one such ecosystem different species and different distributions of plants can service that is often identified as justification for restoring interact with the heterogeneous distribution of resources in wetlands (Osborne & Kovacic 1993; Woltemade 2000; the soil matrix, thereby providing some degree of spatial Rabalais et al. 2002; Zedler 2003; Fennessy & Craft structure in soil biota (Ettema & Wardle 2002). By under- 2011). However, restored wetlands may not perform standing how factors at multiple spatial scales influence equally to their natural counterparts. The effectiveness of wetland functions, particularly C, N, and P biogeochem- restored wetlands in removing excess NO -N depends ical cycling, wetland conservation and restoration can be upon NO -N load, connectivity to adjacent waters, water optimized to improve the overall performance and delivery residence time, soil anaerobiosis, microbial communities, *Corresponding author. Email: johnmarton@gmail.com Present address of John M. Marton: Louisiana Universities Marine Consortium, 8124 Highway 56, Chauvin, LA 70344, USA. © 2014 Taylor & Francis International Journal of Biodiversity Science, Ecosystem Services & Management 37 of ecosystem services (Schlesinger et al. 1996; Bruland structure would be present in restored wetlands. Previous et al. 2006; Grunwald et al. 2007). homogenization of the soil profile via cultivation in the Denitrification can be a significant pathway of N restored wetlands, combined with a relatively short time removal in wetland and riparian buffers, though rates can following restoration, would likely result in lower variabil- differ between natural and restored systems based on ity and spatial structure of denitrification and associated soil factors such as soil organic C content, NO -N concentra- properties relative to the reference depressional wetlands tions, soil texture, and soil moisture (Hanson et al. 1994; (Bruland et al. 2006; Orr et al. 2014). Hunter & Faulkner 2001; Burgin & Hamilton 2007; Orr et al. 2007; Peralta et al. 2010). However, denitrification and the controlling factors are not distributed homoge- Methods nously throughout the wetlands and riparian areas and Site description often exhibit high spatial variability (Ettema et al. 1998; We sampled two natural and two restored wetlands in Bruland & Richardson 2004, 2005; King et al. 2004; Newton County in northwestern Indiana (Figure 1). Bruland et al. 2006). For example, Harms et al. (2009) Restored wetlands were on the Kankakee Sands Prairie found that water vectors from nearby streams influenced Restoration macrosite owned by The Nature Conservancy distributions of C and N in a semiarid floodplain; in one of and were restored in 2001 under the Wetlands Reserve their two sites, C and N spatial patterns were good pre- Program by filling in the surrounding drainage ditches dictors of denitrification. (C. O’Leary, personal communication). Prior to restora- Spatial variability of soil properties and processes can tion, the land was used for row-crop agriculture. The wet- be influenced by biotic and abiotic factors such as vegeta- lands and surrounding prairies are actively managed by tion distributions, soil invertebrate and microbial commu- seeding, invasive removal, and prescribed fires every 2– nities, surface and subsurface hydrology, nutrient loading, 3 years. These wetlands are depressional, precipitation-fed and sedimentation (Ettema et al. 1998; Ettema & Wardle systems, and the soils are mapped as Granby loamy fine 2002; King et al. 2004; Bruland et al. 2006). sand (sandy, mixed, mesic Typic Endoaqualls). The sur- Denitrification in natural and restored riverine and non- rounding land-use is a mixture of restored mesic prairie riverine wetlands in North Carolina was best predicted by and row-crop agriculture. Dominant vegetation in the wet- NO -N, soil organic C, and soil moisture, although the lands consists of Schoenoplectus pungens (Vahl) Palla, strength of relationships varied among sites (Bruland et al. Polygonum pensylvanicum L., Eleocharis erythropoda 2006). Further, both means and variability of NO -N, Steud., Juncus brachycephalus (Engelm.) Buchenau, NH -N, and denitrification differed between restored and Scirpus cyperinus (L.) Kunth, Leersia oryzoides (L.) Sw., natural wetlands. Natural wetlands had greater variability Phalaris arundinacea L., and Solidago altissima L. in NO -N and denitrification than restored or created wet- Natural wetlands were located at the Willow Slough lands, whereas NH -N exhibited similar variability Fish and Wildlife Area owned and managed by the Indiana between sites. Similarly, Gallardo (2003) also observed Department of Natural Resources. Similar to the restored that flooding influences larger scale spatial patterns of wetlands, natural wetlands are depressional, precipitation- soil nutrients in floodplain forests, though vegetation also fed wetlands underlain by Adrian drained muck (sandy, influenced soil properties, particularly C and N, at smaller mixed, euic, mesic Terric Haplosaprists) and dominated by scales. Schlesinger et al. (1996) also found strong spatial Calamagrostis canadensis (Michx.) P. Beauv. and Scirpus patterns in soil N in desert soils, with spatial variation cyperinus (L.) Kunth. resulting from distributions of different shrubs and grasses. Conversely, Burke et al. (1999) showed that topography, not vegetation, influenced soil C pools in a shortgrass Soil sampling steppe, with C pools greater at toeslope positions. Collectively, these processes interact to lead to the creating We established a 32 × 32 m sampling grid in each wetland of microsite variability within the soil profile. in August 2010. Each grid was partitioned into four rows Often, however, comparisons of ecosystem services and four columns for a total of 16 subgrids in each site between natural and restored wetlands ignore their under- (Figure 2). The centroid of the each subgrid was marked lying spatial structure. Failing to take into account the with polyvinyl chloride pipe, and three randomly selected spatial variability of soil properties and processes increases soil cores (8.5 cm diameter × 5-cm deep) were collected the difficulty in identifying factors controlling ecosystem around each centroid for a total of 48 soil cores per wet- services such as C sequestration, nutrient accumulation, and land (192 cores total). Initially, 15-cm-deep cores were denitrification. We investigated the effects of restoration on taken and sectioned into 0- to 5-cm and 5- to 15-cm the spatial variability of denitrification in two restored and sections; however, only the 0- to 5-cm samples were two natural depressional wetlands. Our hypotheses were (1) analyzed to keep the sample size manageable. Cores natural wetlands would have greater variability of denitrifi- were collected with a stainless-steel piston corer, placed cation and associatedsoilpropertiesthanrestoredwetlands, into resealable plastic bags, and returned to the laboratory and (2) denitrification and soil properties would be spatially on ice. All soils were collected over 2 days in late August autocorrelated in natural wetlands, whereas no spatial after at least 1 week of no rain. Restored wetlands were 38 J.M. Marton et al. Newton County, Indiana, USA Legend Natural Wetlands Restored Wetlands G ^ 0 3 6 12 18 24 Kilometers Figure 1. Map of study sites in northwestern Indiana. Polygon represents property boundary of Kankakee Sands Nature Preserve. sampled on the first day and natural wetlands were in mass of a 5-g sample after drying to a constant weight. sampled on the following day. All soil samples were Plant-available N was extracted from soil samples using 2N processed within 48 hr of collection. Following the soil KCl (Mulvaney 1996) and analyzed using a Lachat collection, surface water was collected in 20-L carboys, Quickchem (Lachat Instruments, Loveland, CO, USA). returned to the laboratory, and filtered through 0.20-µm Unamended denitrification was measured with the acety- mesh for use in denitrification assays. lene-inhibition method (Tiedje 1994). Twenty-five grams of field-moist soil were placed into 125-mL Wheaton bottles with screw caps equipped with gray butyl septa. Each bottle Soil analyses received 50 mL of filtered site water amended with chlor- amphenicol (0.21 µM) to inhibit microbial growth. The Field-moist soils were analyzed for soil moisture content, bottles were flushed for 5 minutes with ultra-high-purity plant-available N (NO -N, NH -N), and denitrification. Soil 3 4 N gas, and the headspace was adjusted to 10% atm with moisture content was determined by measuring the change 2 International Journal of Biodiversity Science, Ecosystem Services & Management 39 least in one wetland. Data were tested for normality using Centroid the Kolmogorov–Smirnov test (α = 0.05) and, when neces- Sample sary, were natural-log-transformed to achieve a normal dis- tribution (IBM SPSS, Armonk, NY, USA). Means and standard errors were calculated for each soil property, and the variance between natural and restored wetlands was compared using Levene’stest (α = 0.01). Due to the weak observed autocorrelation, means were compared with a one- way ANOVA with Tukey’s post hoc test. Omnidirectional variograms were generated for denitri- fication and associated soil properties to evaluate their actual spatial structure in each of the four wetlands. Variograms describe how data are correlated with distance, typically in a specified direction; omnidirectional variograms essentially 32 m combine all directional variograms into a single plot, indi- Figure 2. Representative sampling scheme. Three samples were cating overall spatial continuity or structure in empirical collected at random directions and distances up to 4 m from each data. They indicate mean variability between multiple of the 16 centroids for a total of 48 soil cores per site. Lag values pairs of points across all lag distances in each possible ranged from 3.58 to 40.56 m with the number of neighbors direction (Eastman 2009). In order to test for the presence within a lag distance ranging from 2 to 30. of spatial autocorrelation among soil properties in each wet- land, global Moran’s I was calculated. Moran’s I tests for acetylene to block the reduction of N OtoN .Incubations autocorrelation among samples by comparing the variation 2 2 were conducted for 90 minutes at 25°C and a 5-mL sample of paired points within a specified lag distance (spatial was collected after 30, 60, and 90 minutes and stored in 2- covariation) to the total variance (Moran 1950; Legendre mL evacuated Wheaton vials sealed with aluminum crimp & Fortin 1989). The significance of the test is estimated by tops and gray butyl septa. The bottles were vigorously generating a z-score from a randomly permutated distribu- shaken by hand for 30 seconds prior to the collection to tion (Moran 1950; Legendre & Fortin 1989). Values range equilibrate N O between the soil slurry and headspace. from −1 indicating that the neighboring values are dissimilar After collection, a 5-mL mixture of N and acetylene (9:1) (negative spatial autocorrelation) to +1 indicating that neigh- was added to maintain constant pressure. Nitrous oxide boring values are similar (positive spatial autocorrelation). concentrations were measured using a gas chromatograph The Moran’s I test was performed over a range of lag (SRI Instruments, Menlo Park, CA, USA) with an electron distances from the lowest value at each site up to 20 m. capture detector, and concentrations were corrected for dilu- The lowest lag distance is the shortest distance between tion through multiple sample collections. Denitrification points such that each location has at least one neighbor. rates were determined by regressing N O concentrations The global Moran’s I test is a ‘global’ statistic of spatial against time. All denitrification rates were expressed on autocorrelation, meaning that clusters of high values (hot both a dry weight basis by correcting for the soil moisture spots) and low values (cold spots) are not identified (Moran content and an area basis by multiplying dry weight rates by 1950). The Getis–Ord Gi* test (Getis & Ord 1992) was used the sampling depth and bulk density. to detect hot spots of denitrification in each wetland. Hot Following analyses of field-moist soils, remaining spots (and cold spots) are identified as clusters with signifi- soils were dried, ground, and passed through a 2-mm cantly greater (or lower) values relative to surrounding mesh sieve to determine bulk density, organic C, and points. Omnidirectional variograms were generated using total N. Carbonates were removed by placing subsamples IDRISI Taiga (Clark Labs, Worcester, MA, USA), and all in a dessicator with a beaker of concentrated HCl for 24 hr other geostatistical analyses were performed using GeoDa (Hedges & Stern 1984). Organic C and total N were (Luc Anselin, University of Illinois, Urbana-Champaign, determined from these subsamples using a Perkin-Elmer Urbana, IL, USA; Anselin et al. 2006). 2400 CHN Analyzer (Perkin-Elmer, Waltham, MA, USA). Bulk density was calculated by dividing the total dry Results weight of the soil sample by the volume of the core (Blake & Hartge 1986). All results were expressed on a Denitrification was greater in restored wetlands (30 ng −1 −1 dry gram basis. NOg hr ) than in natural wetlands (1.1 ng −1 −1 N Og hr ) and was most strongly correlated to C:N ratios (rho = 0.50, p < 0.01). Due to the differences in bulk Geostatistics and data analysis density between the natural and restored wetlands, denitri- A priori, soil cores within each wetland were not considered fication was corrected for the specific bulk density and independent samples due to their close proximity to one sampling depth to express rates based on area. When another (Bruland et al. 2006). All variables exhibited weak expressed on a mass basis, the denitrification between the but significant positive spatial autocorrelation (see below) at two natural wetlands significantly differed (Figure 3a); 32 m 40 J.M. Marton et al. restored wetlands regardless of units. Because rates and A B B patters were comparable regardless of units, rates based on the dry weight were used for geostatistical analysis. Restored wetlands also had greater bulk density and C:N ratios, whereas natural wetlands had greater soil moisture, plant-available N (NO -N, NH -N), organic 3 4 C, and total N (Table 1). Levene’stest(α =0.01) indi- cated significantly greater variation in denitrification, soil moisture, bulk density, total N, and C:N ratios (Table 1) in restored wetlands than in natural wetlands. –2 Natural wetlands, on the other hand, had greater var- –4 iance of NO -N. Plant-available NH -N and organic C 3 4 had comparable variation between natural and restored 10 wetlands. B C A C Overall, omnidirectional variograms did not indicate marked spatial trends in denitrification or other measured variables across the individual natural or restored wetlands (data not shown). Global Moran’s I analyses revealed a wide range of spatial autocorrelation among soil properties and the four wetlands, at variable lag distances. Denitrification exhibited positive spatial autocorrelation in both natural wetlands between 4 and 5 m and up to 16 m in restored wetland 1; no spatial autocorrelation was –2 detected at any lag distance in restored wetland 2 (Table 2, –4 Figure 4). Organic C was autocorrelated in restored wet- 1 2 1 2 land 2 only up to 20 m, whereas no spatial autocorrelation was found in natural wetlands or restored wetland 1 (Table 2, Figure 5). Conversely, NO -N was spatially Figure 3. Boxplots of denitrification (natural-log transformed) 3 from the two natural and restored wetlands (n = 48). Different autocorrelated in both natural wetlands, but only in one letters indicate significant differences based on Tukey’s test. of the two restored wetlands (Table 2). The two natural Variance was significantly greater in restored wetlands based on wetlands exhibited similar patterns for both denitrification Levene’s test (p < 0.05). and soil organic C (Figures 4 and 5, respectively). However, denitrification in restored wetland 2 had a com- parable pattern to the natural wetlands (Figure 4), whereas however, no differences were detected between these two soil organic C in restored wetland 1 mimicked the natural sites when expressed on an area basis (Figure 3b). wetlands (Figure 5). Denitrification patterns were comparable between the two Table 1. Site means (±1 SE) of soil properties at each of the sampled wetlands. Moisture −3 −1 −1 (%) Bulk density (g cm )NO -N (µg g )NH -N (µg g ) Organic C (%) Total N (%) C:N ratio (mol:mol) 3 4 Restored 1 36 ± 1.7 0.83 ± 0.03 4.6 ± 0.9 0.86 ± 0.14 2.5 ± 0.31 0.15 ± 0.03 22.6 ± 1.2 Restored 2 11 ± 0.9 1.1 ± 0.03 3.8 ± 1.0 2.0 ± 0.35 1.9 ± 0.13 0.16 ± 0.01 14.2 ± 0.11 Natural 1 47 ± 1.2 0.35 ± 0.02 26 ± 2 5.9 ± 0.65 16 ± 1 1.5 ± 0.08 12.5 ± 0.07 Natural 2 46 ± 1.4 0.40 ± 0.02 79 ± 7 1.8 ± 0.20 16 ± 1 1.3 ± 0.08 13.7 ± 0.12 Mean a a b a a Restored 23 ± 1.6 0.97 ± 0.03 4.1 ± 0.7 1.4 ± 0.19 2.2 ± 0.17 0.15 ± 0.01 18.4 ± 0.75 b b a b b Natural 46 ± 1.0 0.37 ± 0.02 53 ± 4 3.8 ± 0.40 16 ± 1 1.4 ± 0.06 13.1 ± 0.09 Note: Different letters indicate significantly different variances between natural and restored wetlands based on Levene’s test (α = 0.01). Table 2. Lag distances (m) with significant spatial autocorrelation in the two natural and two restored depressional wetlands based on Moran’s I analysis. Denitrification Bulk density Soil moisture NO -N NH -N Organic C Total N C:N 3 4 Restored wetland 1 16 20 – – 5.8 –– – Restored wetland 2 – 7.6 20 12.1 – 20 20 9.1 Natural wetland 1 5.1 – 20 5.1 –– – – Natural wetland 2 4.2 –– 5.8 7.7 –– – Note: No significant spatial autocorrelation present at any lag distance. –2 –1 –1 –1 Denitrification (ng N O cm hr ) Denitrification (ng N O g hr ) 2 2 (LN-Transformed) (LN-Transformed) International Journal of Biodiversity Science, Ecosystem Services & Management 41 Denitrification Soil Organic C 0.4 0.5 Restored 1 Restored 1 Restored 2 0.4 Restored 2 Natural 1 Natural 1 0.3 Natural 2 0.3 Natural 2 0.2 0.2 0.1 0.1 0.0 –0.1 0.0 –0.2 –0.1 –0.3 2 4 6 8 10 12 14 16 18 20 2468 10 12 14 16 18 20 Lag (m) Lag (m) Figure 4. Moran’s I values for denitrification across lag dis- Figure 5. Moran’s I values for soil organic C across lag dis- tances in natural and restored wetlands. tances in natural and restored wetlands. Getis–Ord Gi* results indicated significant hot spots in natural wetland 2. Hot spots and cold spots were also of denitrification in all four wetlands (Figure 6). Restored identified for other measured soil properties, though wetland 1 had 6 significant hot spots and restored wet- denitrification hot spots did not correspond to those land 2 only had one. Six hot spots were detected in of NO -N, soil moisture, or organic C in any of the natural wetland 1 and only three hot spots were detected wetlands. (a) (b) Restored 2 Restored 1 Not Not Significant Significant Low Low High High 0 8 16 24 32 0 8 16 24 32 X Distance (m) X Distance (m) Natural 2 Natural 1 (c) (d) Not Not 32 32 Signficant Significant Low Low High High 24 24 0 0 0 8 16 24 32 0 8 16 24 32 X Distance (m) X Distance (m) Figure 6. Denitrification hot spots and cold spots in the (a) restored wetland 1, (b) restored wetland 2, (c) natural wetland 1, and (d) natural wetland 2 based on the Getis–Ord Gi* test (α = 0.05). Moran's I Y Distance (m) Y Distance (m) Y Distance (m) Y Distance (m) Moran's I 42 J.M. Marton et al. Discussion For most soil properties, restored wetlands had greater variance (Table 1) and greater significant lag distances Higher denitrification in restored wetlands relative to nat- (Table 2), though causal processes behind the observed ural wetlands was an unexpected result. Other studies have spatial trends are not entirely clear. Bruland and found greater denitrification in natural wetlands than in Richardson (2005) also measured significant spatial trends restored wetlands in multiple hydrogeomorphic settings. in created, restored, and natural riverine and non-riverine Hunter and Faulkner (2001) measured greater denitrification −1 −1 wetlands. They suggested that overbank flooding from the rates in natural bottomland wetlands (657 ng N Og hr ) adjacent river could have led to increasing soil organic compared to restored bottomland wetlands (167 ng −1 −1 matter and decreasing sand content with increasing dis- N Og hr ), and Bruland et al. (2006) found greater tance from the stream. Bruland and Richardson (2005) denitrification in natural riverine and non-riverine wetlands also found linear and non-linear trends in soil properties compared to created and restored wetlands. In both studies, from non-riverine wetlands, which they hypothesized to be positive correlations were detected between denitrification related to variations in unmeasured properties such as and soil moisture. Using a larger set of wetlands (n =20), topography, above- and belowground biomass, and soil including the four used in this study, Marton et al. (2014) biota (Schlesinger et al. 1996; Ettema & Wardle 2002). measured greater ambient and potential denitrification in These study sites were depressional systems, likely having natural wetlands than in restored wetlands, with denitrifica- minimal horizontal movement of surface water. In contrast tion rates positively correlated to soil NO -N and moisture. to our study, Bruland et al. (2006) found lower variation in In the current study, denitrification was positively correlated denitrification and NO -N in restored non-riverine wet- to C:N ratios (rho = 0.50, p < 0.01) and negatively corre- lands compared to natural wetlands. Conversely, they lated to both NO -N (rho = −0.50, p < 0.01) and soil found equal or greater variation of soil moisture and organic C (rho = −0.34, p < 0.01). One possible explanation NH -N in restored non-riverine wetlands. Their restored is that the N O produced during incubations in the current wetlands were former agricultural fields, and they reported study was the result of dissimilatory nitrate reduction to that surface soils were homogenized during restoration. ammonia (DNRA) rather than denitrification, which can However, their results were not consistent between river- occur in substrates with high C:N ratios and produces ine and non-riverine wetlands. In one of their riverine N O as a by-product (Burgin & Hamilton 2007). In created sites, spatial variability of denitrification was greater in freshwater wetlands in Texas, Scott et al. (2008)found the restored wetland than in the natural wetland. Orr et al. DNRA occurring simultaneously with denitrification, (2014) also found greater denitrification rates in a restored though DNRA frequently accounted for a small portion floodplain wetland relative to a natural floodplain wetland. (<5%) of total NO -N removal. However, they did find They found greater variability and no spatial structure in that DNRA accounted for up to 36% of NO -N removal denitrification at the restored site, consistent with our in one location characterized by low overlying water NO - findings. However, denitrification exhibited a strong spa- N concentrations. In a review of aquatic N-cycling pro- tial structure in their natural site. The lack of a significant cesses, Burgin and Hamilton (2007) discuss the variability spatial structure in the current study suggests that means of DNRA in total NO -N removal across freshwater and and variances of denitrification and soil properties across marine systems, and that this process may be less important the sampled natural and restored wetlands are not spatially in freshwater wetlands. The magnitude of DNRA in the dependent or vary at spatial scales greater or smaller than current study, while not known, could have implications for the scale at which these measurements were made. the overall delivery of ecosystem services. If, for example, Another possibility is the complexity of wetland eco- DNRA is a more important NO -N removal pathway than systems, with intra-site variability, soil and hydrological denitrification, then N O production may exceed N pro- 2 2 dynamics, and disturbance regimes make it difficult to duction, which is a less favorable outcome as N Ois a accurately parse out the spatial variability in these soil highly potent greenhouse gas. properties and processes. Several factors, alone or acting Soil organic C was greater in natural wetlands relative in concert, can influence the spatial distribution and struc- to restored wetlands. Development of soil organic matter is ture of soil properties, which in turn can influence the a slow process (Craft et al. 1999; Ballantine & Schneider spatial structure and magnitude of denitrification. 2009), and restored wetlands in the current study were only Following the end of cultivation, vegetation communities 10 years old. Further, the prairies surrounding the restored can go through successional stages, which in turn can wetlands are burned every 2–3 years. The high sand content influence soil properties such as C and N pools, soil of the restored wetlands (91% sand, Marton et al. 2014) moisture, oxygen content, and ultimately denitrification promotes drainage to groundwater, thereby allowing fires to (Gross et al. 1995; Rotkin-Ellman et al. 2004; Diekmann burn through the wetlands, which may inhibit the buildup et al. 2007). Bachand and Horne (1999) found greater of soil organic C (Neff et al. 2005). Anderson et al. (2005) rates of NO -N removal in a constructed wetland with a found that soil organic matter concentrations increased over mixed vegetation community of Typha spp. and Scirpus 10 years in two created riverine marshes in Ohio, while also spp. relative to individual monocultures. Gross et al. increasing in spatial variability due to variability in inunda- (1995) found greater spatial variability in soil N in mid- tion, vegetation, and sediment deposition. International Journal of Biodiversity Science, Ecosystem Services & Management 43 successional fields relative to a newly abandoned field and 2003). Important factors controlling denitrification rates a forest, while Rotkin-Ellman et al. (2004) showed that are NO -N, labile C, and anoxic conditions (Hunter & different tree species lead to differences in soil organic Faulkner 2001; McClain et al. 2003; Bruland et al. 2006; matter patches, which can be hot spots of denitrification. Ullah & Faulkner 2006; Bruland et al. 2009). Generally, Soil texture between the natural and restored wetlands in hot spots of denitrification and controlling factors should the current study was relatively comparable (Marton et al. correspond. In the current study, however, denitrification 2014), though there soil organic matter was greater in was not correlated with NO -N, soil organic C, or soil natural wetlands relative to restored wetlands. Though moisture. Further, denitrification hot spots did not overlap not statistically greater, the restored wetlands had greater with variables required for denitrification (e.g., soil moist- plot scale (1 m ) and site-level species richness relative to ure, NO -N, organic C), suggesting that other factors may the natural wetlands (Hopple & Craft 2013). The greater have been influencing denitrification. Plausible factors that plant diversity, coupled with differences in organic matter may explain denitrification include differences in micro- and slight variations in soil texture, could have led to the bial communities, soil texture, or interactions between observed differences in magnitude and spatial structure of NO -N, organic C, and moisture (Hanson et al. 1994; denitrification and other soil properties (e.g., soil organic Bruland et al. 2006; Peralta et al. 2010). Also, because C) between and within natural and restored wetlands. denitrification is a microbial process, our scale of measure- The observed spatial variance in soil properties from ment at the nearest tenth of a meter may have been too natural and restored wetlands was not entirely the result of coarse to detect relationships between denitrification and random processes. Significant positive spatial autocorrela- associated soil properties. tion for denitrification and several other soil properties Other studies have focused on identifying hot spots of (Figures 4 and 5, Table 2) suggested that, in addition to denitrification and soil properties, though hot spots were the plot-scale directional trends, significant spatial variabil- identified visually either based on interpolated maps or ity was present as well. Though the Moran’s I analysis based on parametric statistics that ignore spatial autocor- shows whether variables are globally autocorrelated over relation among samples (Rotkin-Ellman et al. 2004; space, the test does not provide insight into the causality of Bruland & Richardson 2005; Bruland et al. 2006). Our the spatial patterns. Future work could incorporate greater study differed in that we used the Getis–Ord Gi* test to sample sizes and attempt explanatory models, such as spa- statistically verify hot spots of denitrification. Ignoring the tially weighted regressions. Spatial patterns in vegetation, spatial aspect of soil properties and processes in attempt- hydrodynamics, and soil biota have been found to influence ing to estimate hot spots artificially inflates the signifi- soil nutrient pools and processes (Schlesinger et al. 1996; cance of the statistical tests by treating each sample as an Ettema & Wardle 2002;King et al. 2004). Schlesinger et al. independent observation, though due to their proximity in (1996) found that available N (NO -N, NH -N) was space, the samples cannot truly be considered independent 3 4 strongly autocorrelated within 20 cm of the perennial (Moran 1950; Legendre & Fortin 1989; Goovaerts 1998). bunchgrass Bouteloua eriopoda in desert soils, resulting The comparable, and in some cases greater, spatial from belowground nutrient cycling by the shrubs and asso- variability in denitrification and soil properties in restored ciated soil microorganisms. In a restored riparian wetland in wetlands relative to natural wetlands was surprising. Georgia, Ettema et al. (1998) found that soil properties Previous research has suggested that prior land use affects (NO -N, soil moisture) exhibited strong spatial autocorrela- soil properties, particularly agriculture that homogenizes tion. Plant-available NO -N was autocorrelated at distances soil structure both vertically and horizontally (Bruland up to 84 m, whereas in the current study, NO -N was et al. 2003). We found the opposite trend in that denitrifi- autocorrelated at distances up to 8 m. One possible expla- cation in restored wetlands exhibited stronger spatial varia- nation for the large discrepancy between distances in the bility and a comparable numbers of hot spots relative to two studies is the influence of hydrology. Ettema et al. natural wetlands. These results suggest that, despite prior (1998) measured soil variables in a riparian wetland that cultivation and homogenization of soil profiles, restoration received periodic pulses of water from the adjacent river, can result in systems with comparable ranges of biogeo- distributing sediment and nutrients over greater distances. chemical functioning. Conversely, our study sites were precipitation-fed depres- There are both advantages and disadvantages of our sional wetlands, which received little in the way of sampling design. Our sample plots were only 32 × 32 m allochthonous inputs. King et al. (2004)measured the var- which prevented us from adequately sampling the entire iation and relationships between environmental factors and wetland and thereby capturing site-scale variability. vegetation communities along a 10-km transect of wetlands However, we were able to sample four wetlands in total in the Everglades. They found that nutrients (K, N, and P) allowing us to make comparisons both between and within and hydrology influenced vegetation patterns at the wetland natural and restored depressional wetlands. For example, scale, whereas P loading controlled vegetation community though spatial patterns of denitrification and soil organic C patterns along the 10-km transect. were comparable between the two natural wetlands, the Denitrification hot spots occur where reactants and patterns differed between the two restored sites. Though it conditions suitable for denitrification are disproportio- cannot be determined why these differences exist, it is nately higher than the surrounding areas (McClain et al. possible that differences in plant density and diversity 44 J.M. Marton et al. between the two restored sites are influencing below- of Agriculture, Natural Resources Conservation Service, Conservation Effects Assessment Program through The Great ground dynamics. Lakes-Northern Forest Cooperative Ecosystem Studies Unit, With a smaller extent and finer grain, we were able to Cooperative Agreement Number 68-7482-9-516. investigate patterns in denitrification and the associated soil properties over shorter spatial scales. Changing either the extent or the grain of the study area would ultimately change the variability, and thus the predictability, of a References given parameter (Wiens 1989). This combined with the Anderson C, Mitsch W, Nairn R. 2005. Temporal and spatial fact that we were unable to fit a reliable semivariogram to development of surface soil conditions at two created river- any of our variables implies an inherent difficulty and ine marshes. J Environ Qual. 34:2072–2081. Anselin L, Syabri I, Kho Y. 2006. GeoDa: an introduction to potential inapplicability in scaling up our results to the spatial data analysis. Geogr Anal. 38:5–22. wetland scale. Though these results may not be effectively Bachand P, Horne A. 1999. Denitrification in constructed free- scaled to the wetland scale, findings from this and other water surface wetlands: II. Effects of vegetation and tem- similar studies (Bruland & Richardson 2004, 2005; Orr perature. Ecol Eng. 14:17–32. et al. 2014) can provide useful information guiding wet- Ballantine K, Schneider R. 2009. Fifty-five years of soil devel- opment in restored freshwater depressional wetlands. Ecol land restoration and post-restoration monitoring and Appl. 19:1467–1480. assessment. For example, reestablishment of hydrology, Blake G, Hartge K. 1986. Bulk density. In: Klute A, editor. native vegetation, and organic matter amendments will Methods of soil analysis, part 1. Physical and mineralogical help a restored system have comparable or greater deni- methods: agronomy monograph no. 9. 2nd ed. Madison trification potential when compared to natural reference (WI): American Society of Agronomy; p. 363–375. Bruland G, Hanchey M, Richardson C. 2003. Effects of agricul- sites. Further, these results also show that a high number ture and wetland restoration on hydrology, soils, and water of samples can provide a good estimation of the variability quality of a Carolina bay complex. Wetl Ecol Manage. of soil properties and processes, though the sampling 11:141–156. extent would have to cover the majority of the wetland Bruland G, Richardson C. 2004. A spatially explicit investigation to fully understand the spatial structure and to appropri- of phosphorus sorption and related soil properties in two riparian wetlands. J Environ Qual. 33:785–794. ately scale those findings. While it is not feasible to con- Bruland G, Richardson C. 2005. Spatial variability of soil proper- duct spatially intensive investigations of denitrification in ties in created, restored, and paired natural wetlands. Soil Sci most cases, more studies that are conducted with a spatial Soc Am J. 69:273–284. sampling design will increase our understanding of the Bruland G, Richardson C, Daniels W. 2009. Microbial and controls on spatial structuring of ecosystem services and geochemical responses to organic matter amendments in a created wetland. Wetlands. 29:1153–1165. how this structure changes over time. Bruland G, Richardson C, Whalen S. 2006. Spatial variability of Incorporating spatial patterns provides greater under- denitrification potential and related soil properties in created, standing of how individual and multiple environmental restored, and paired natural wetlands. Wetlands. 26:1042–1056. factors influence the distribution and development of soil Burgin A, Hamilton S. 2007. Have we overemphasized the role properties and processes. This is particularly true for of denitrification in aquatic ecosystems? A review of nitrate removal pathways. Front Ecol the Environ. 5:89–96. restoring wetlands, in which the primary goal is to return Burke I, Lauenroth W, Riggle R, Brannen P, Madigan B, Beard some ecosystem service to the landscape. Our results S. 1999. Spatial variability of soil properties in the showed that, despite lower concentrations of soil C and Shortgrass Steppe: the relative importance of topography, N, restored wetlands had more complex spatial variability grazing, microsite, and plant species in controlling spatial in denitrification, soil moisture, bulk density, total N, and patterns. Ecosystems. 2:422–438. Craft C, Reader J, Sacco J, Broome S. 1999. Twenty-five years C:N ratios compared to natural wetlands and provided a of ecosystem development of constructed Spartina alterni- greater range of biogeochemical transformations. flora (Loisel) marshes. Ecol Appl. 9:1405–1419. Increased knowledge of the spatial structure of soil proper- Diekmann L, Lawrence D, Okin G. 2007. Changes in the spatial ties will further reveal factors controlling processes such variation of soil properties following shifting cultivation in a as denitrification at multiple spatial scales in restored wet- Mexican tropical dry forest. Biogeochemistry. 84:99–113. Eastman J. 2009. IDRISI Taiga: guide to GIS and image proces- lands. By understanding the underlying factors driving sing. Worcester (MA): Clark Labs, Clark University. denitrification and other soil processes (e.g., accretion, Ettema C, Coleman D, Vellidis G, Lowrance R, Rathbun S. 1998. sedimentation, P sorption), wetland restoration can be Spatiotemporal distributions of bacterivorous nematodes and improved to maximize the delivery of ecosystem services. soil resources in a restored riparian wetland. Ecology. 79:2721–2734. Ettema C, Wardle D. 2002. Spatial soil ecology. Trends Ecol Evol. 17:177–183. Acknowledgments Fennessy S, Craft C. 2011. Agricultural conservation practices We would like to thank Chip O’Leary and Stephanie Frischie at increase wetland ecosystem services in the Glaciated Interior The Nature Conservancy at the Kankakee Sands Preserve and the Plains. Ecol Appl. 21:S49–S64. Indiana Department of Natural Resources for access to their Gallardo A. 2003. Spatial variability of soil properties in a flood- property. We would also like to thank Ellen Herbert, Anya plain forest in Northwest Spain. Ecosystems. 6:564–576. Hopple, and Bri Richards for help with the collection, prepara- Getis A, Ord J. 1992. The analysis of spatial association by use tion, and analysis of samples, and Maggie Marton for editorial of distance statistics. Geogr Anal. 24:189–206. assistance. This study was funded by the US Department International Journal of Biodiversity Science, Ecosystem Services & Management 45 strategies to counter a persistent ecological problem. Gittins R. 1968. Trend-surface analysis of ecological data. J Ecol. BioScience. 51:373–388. 56:845–869. Moran P. 1950. Notes on continuous stochastic phenomena. Goovaerts P. 1998. Geostatistical tools for characterizing the Biometrika. 37:17–23. spatial variability of microbiological and physico-chemical Mulvaney R. 1996. Nitrogen- inorganic forms. In: Sparks DL, soil properties. Biol Fertil Soils. 27:315–334. editor. Methods of soil analysis, part 3: chemical methods. Gross K, Pregitzer K, Burton A. 1995. Spatial variation in nitro- Madison (WI): Soil Science Society of America; p. 1123– gen availability in three successional plant communities. J Ecol. 83:357–367. Neff J, Harden J, Gleixner G. 2005. Fire effects on soil organic Grunwald S, Reddy K, Newman S, DeBusk W. 2004. Spatial matter content, composition, and nutrients in boreal interior variability, distribution and uncertainty assessment of soil Alaska. Can J For Res. 35:2178–2187. phosphorus in a south Florida wetland. Environmetrics. Orr C, Stanley E, Wilson K, Finlay J. 2007. Effects of restoration 15:811–825. and reflooding on soil denitrification in a leveed Midwestern Grunwald S, Reddy K, Prenger J, Fisher M. 2007. Modeling of floodplain. Ecol Appl. 17:2365–2376. the spatial variability of biogeochemical soil properties in a freshwater ecosystem. Ecol Modell. 201:521–535. Orr CH, Predick KI, Stanley EH, Rogers KL. 2014. Spatial Hanson G, Groffman P, Gold A. 1994. Denitrification in riparian autocorrelation of denitrification in a restored and a natural wetlands receiving high and low groundwater nitrate inputs. floodplain. Wetlands. 34:89–100. J Environ Qual. 23:917–922. Osborne L, Kovacic D. 1993. Riparian vegetated buffer strips in Harms TK, Wentz EA, Grimm NB. 2009. Spatial heterogeneity water-quality restoration and stream management. Freshw of denitrification in semi-arid floodplains. Ecosystems. Biol. 29:243–258. 12:129–143. Peralta A, Matthews J, Kent A. 2010. Microbial community Hedges J, Stern J. 1984. Carbon and nitrogen determinations of structure and denitrification in a wetland mitigation bank. carbonate-containing solids. Limnol Oceanogr. 29:657– Appl Environ Microbiol. 76:4207–4215. 663. Rabalais NN, Turner RE, Scavia D. 2002. Beyond science into policy: Gulf of Mexico hypoxia and the Mississippi River. Hopple A, Craft C. 2013. Managed disturbance enhances biodi- BioScience. 52:129–142. versity of restored wetlands in the agricultural Midwest. Ecol Rotkin-Ellman M, Addy K, Gold A, Groffman P. 2004. Tree Eng. 61:505–510. species, root decomposition and subsurface denitrification Hunter R, Faulkner S. 2001. Denitrification potentials in restored potential in riparian wetlands. Plant Soil. 263:335–344. and natural bottomland hardwood wetlands. Soil Sci Soc Am Schlesinger W, Raikes J, Hartley A, Cross A. 1996. On the J. 65:1865–1872. spatial pattern of soil nutrients in desert ecosystems. King R, Richardson C, Urban D, Romanowicz E. 2004. Spatial Ecology. 77:364–374. dependency of vegetation-environment linkages in an anthro- Scott JT, McCarthy MJ, Gardner WS, Doyle RD. 2008. pogenically influenced wetland ecosystem. Ecosystems. Denitrification, dissimilatory nitrate reduction to ammonium, 7:75–97. and nitrogen fixation along a nitrate concentration gradient in Kovacic DA, Twait RM, Wallace MP, Bowling JM. 2006. Use of a created freshwater wetland. Biogeochemistry. 87:99–111. created wetlands to improve water quality in the Midwest- Tiedje J. 1994. Denitrifier enzyme activity. In: Weaver RW, Angle Lake Bloomington case study. Ecol Eng. 28:258–270. S, Bottomley P, editors. Methods of soil analysis, part 2: Legendre P, Fortin M. 1989. Spatial pattern and ecological ana- microbiological and biochemical properties. Madison (WI): lysis. Vegetatio. 80:107–138. Soil Science Society of America; p. 256–257. Marton JM, Fennessy MS, Craft CC. 2014. Functional differ- Ullah S, Faulkner S. 2006. Denitrification potential of different ences between natural and restored wetlands in the Glaciated Interior Plains. J Environ Qual. 43:409–417. land-use types in an agricultural watershed, lower McClain M, Boyer E, Dent C, Gergel S, Grimm N, Groffman P, Mississippi valley. Ecol Eng. 28:131–140. Hart S, Harvey J, Johnston C, Mayorga E, et al. 2003. Wiens J. 1989. Spatial scaling in ecology. Funct Ecol. 3:385–397. Biogeochemical hot spots and hot moments at the interface Woltemade C. 2000. Ability of restored wetlands to reduce of terrestrial and aquatic ecosystems. Ecosystems. 6:301–312. nitrogen and phosphorus concentrations in agricultural drai- Mitsch W, Day J, Wendell Gilliam J, Groffman P, Hey D, nage water. J Soil Water Conserv. 55:303–309. Randall G, Wang N. 2001. Reducing nitrogen loading to Zedler JB. 2003. Wetlands at your service: reducing impacts of the Gulf of Mexico from the Mississippi river basin: agriculture at the watershed scale. Front Ecol Environ. 1:65–72. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Biodiversity Science, Ecosystem Services & Management Taylor & Francis

A comparison of the spatial variability of denitrification and related soil properties in restored and natural depression wetlands in Indiana, USA

Loading next page...
 
/lp/taylor-francis/a-comparison-of-the-spatial-variability-of-denitrification-and-related-6IwQ9qR8Og

References (52)

Publisher
Taylor & Francis
Copyright
© 2014 Taylor & Francis
ISSN
2151-3732
eISSN
2151-3740
DOI
10.1080/21513732.2014.950981
Publisher site
See Article on Publisher Site

Abstract

International Journal of Biodiversity Science, Ecosystem Services & Management, 2015 Vol. 11, No. 1, 36–45, http://dx.doi.org/10.1080/21513732.2014.950981 A comparison of the spatial variability of denitrification and related soil properties in restored and natural depression wetlands in Indiana, USA a b a John M. Marton *, Rinku Roy Chowdhury and Christopher B. Craft School of Public and Environmental Affairs, Indiana University, 702 North Walnut Grove Ave, Bloomington, IN 47405, USA; Department of Geography, Indiana University, 701 East Kirkwood Ave, Bloomington, IN 47405, USA We characterized and compared the variability and spatial patterns of denitrification and soil properties in two natural and two restored depressional wetlands in northern Indiana, USA. Soil properties included soil moisture content, bulk density, plant-available N (NO -N, NH -N), soil organic carbon (C), total nitrogen (N), and C:N ratio. Restored wetlands had greater 3 4 denitrification and higher spatial variability than natural wetlands. Further, restored wetlands had greater bulk density and C: N ratios. In contrast, natural wetlands had greater soil moisture, plant-available N, organic C, and total N. Similar to denitrification, soil moisture, bulk density, NO -N, total N, and C:N ratios had greater variance in restored wetlands than in natural wetlands. Denitrification and several soil properties exhibited positive global spatial autocorrelation, though trends differed between the individual wetlands for soil properties. Variogram analysis suggested little spatial structure in variables at the chosen observational scale. Denitrification hot spots were detected in natural and restored wetlands, though these hot spots did not correspond to hot spots of any of the other soil variables. Overall, spatial patterns of denitrification and soil properties differed between natural and restored wetlands and should be considered when assessing effectiveness of restored wetlands at providing ecosystem services, such as N removal and C storage. Keywords: wetland restoration; denitrification; spatial variability; soil organic carbon; spatial autocorrelation Introduction and soil texture (Mitsch et al. 2001; Bruland et al. 2003, 2006; Ullah & Faulkner 2006). Wetlands exist at the interface of upland and aquatic The spatial structure of soil properties and processes systems and provide many ecosystem services such as adds an additional level of complexity to the comparison carbon (C) sequestration, nutrient (nitrogen [N] and phos- of natural and restored wetlands. Spatial heterogeneity can phorus [P]) retention and transformation, water storage, influence above- and belowground biodiversity, which in and biodiversity support (Mitsch et al. 2001; Zedler turn can affect biogeochemical cycling of C and N 2003; Fennessy & Craft 2011). Wetlands in agricultural (Schlesinger et al. 1996; Ettema et al. 1998; Ettema & landscapes are uniquely positioned to intercept nutrients Wardle 2002). Wetland functions also can exhibit strong from agricultural fields, limiting the potential for down- spatial patterns over multiple scales, ranging from centi- stream eutrophication. In particular, wetlands are effective meters to kilometers (Ettema et al. 1998; King et al. 2004). at removing excess NO -N from surface and groundwater Determining the spatial variation in the nutrient status and draining agricultural fields (Woltemade 2000; Mitsch et al. biogeochemical processes in wetlands will allow managers 2001; Zedler 2003; Kovacic et al. 2006). to determine whether certain locations of a wetland are Wetland restoration aims to reintroduce ecosystem ser- more likely to be a nutrient source or sink (Grunwald et al. vices to the landscape. This is particularly true in the 2004, 2007). Hydrologic flowpaths into and along the agricultural landscapes of Midwestern United States wetlands influence nutrient, particularly NO -N, delivery where extensive wetland loss has occurred due to the to macrophytes and above- and belowground production, expansion of row-crop agriculture (Mitsch et al. 2001; which in turn affects distribution of soil organic matter and Zedler 2003; Fennessy & Craft 2011). Water quality microbial communities (Anderson et al. 2005). Further, improvement through N removal is one such ecosystem different species and different distributions of plants can service that is often identified as justification for restoring interact with the heterogeneous distribution of resources in wetlands (Osborne & Kovacic 1993; Woltemade 2000; the soil matrix, thereby providing some degree of spatial Rabalais et al. 2002; Zedler 2003; Fennessy & Craft structure in soil biota (Ettema & Wardle 2002). By under- 2011). However, restored wetlands may not perform standing how factors at multiple spatial scales influence equally to their natural counterparts. The effectiveness of wetland functions, particularly C, N, and P biogeochem- restored wetlands in removing excess NO -N depends ical cycling, wetland conservation and restoration can be upon NO -N load, connectivity to adjacent waters, water optimized to improve the overall performance and delivery residence time, soil anaerobiosis, microbial communities, *Corresponding author. Email: johnmarton@gmail.com Present address of John M. Marton: Louisiana Universities Marine Consortium, 8124 Highway 56, Chauvin, LA 70344, USA. © 2014 Taylor & Francis International Journal of Biodiversity Science, Ecosystem Services & Management 37 of ecosystem services (Schlesinger et al. 1996; Bruland structure would be present in restored wetlands. Previous et al. 2006; Grunwald et al. 2007). homogenization of the soil profile via cultivation in the Denitrification can be a significant pathway of N restored wetlands, combined with a relatively short time removal in wetland and riparian buffers, though rates can following restoration, would likely result in lower variabil- differ between natural and restored systems based on ity and spatial structure of denitrification and associated soil factors such as soil organic C content, NO -N concentra- properties relative to the reference depressional wetlands tions, soil texture, and soil moisture (Hanson et al. 1994; (Bruland et al. 2006; Orr et al. 2014). Hunter & Faulkner 2001; Burgin & Hamilton 2007; Orr et al. 2007; Peralta et al. 2010). However, denitrification and the controlling factors are not distributed homoge- Methods nously throughout the wetlands and riparian areas and Site description often exhibit high spatial variability (Ettema et al. 1998; We sampled two natural and two restored wetlands in Bruland & Richardson 2004, 2005; King et al. 2004; Newton County in northwestern Indiana (Figure 1). Bruland et al. 2006). For example, Harms et al. (2009) Restored wetlands were on the Kankakee Sands Prairie found that water vectors from nearby streams influenced Restoration macrosite owned by The Nature Conservancy distributions of C and N in a semiarid floodplain; in one of and were restored in 2001 under the Wetlands Reserve their two sites, C and N spatial patterns were good pre- Program by filling in the surrounding drainage ditches dictors of denitrification. (C. O’Leary, personal communication). Prior to restora- Spatial variability of soil properties and processes can tion, the land was used for row-crop agriculture. The wet- be influenced by biotic and abiotic factors such as vegeta- lands and surrounding prairies are actively managed by tion distributions, soil invertebrate and microbial commu- seeding, invasive removal, and prescribed fires every 2– nities, surface and subsurface hydrology, nutrient loading, 3 years. These wetlands are depressional, precipitation-fed and sedimentation (Ettema et al. 1998; Ettema & Wardle systems, and the soils are mapped as Granby loamy fine 2002; King et al. 2004; Bruland et al. 2006). sand (sandy, mixed, mesic Typic Endoaqualls). The sur- Denitrification in natural and restored riverine and non- rounding land-use is a mixture of restored mesic prairie riverine wetlands in North Carolina was best predicted by and row-crop agriculture. Dominant vegetation in the wet- NO -N, soil organic C, and soil moisture, although the lands consists of Schoenoplectus pungens (Vahl) Palla, strength of relationships varied among sites (Bruland et al. Polygonum pensylvanicum L., Eleocharis erythropoda 2006). Further, both means and variability of NO -N, Steud., Juncus brachycephalus (Engelm.) Buchenau, NH -N, and denitrification differed between restored and Scirpus cyperinus (L.) Kunth, Leersia oryzoides (L.) Sw., natural wetlands. Natural wetlands had greater variability Phalaris arundinacea L., and Solidago altissima L. in NO -N and denitrification than restored or created wet- Natural wetlands were located at the Willow Slough lands, whereas NH -N exhibited similar variability Fish and Wildlife Area owned and managed by the Indiana between sites. Similarly, Gallardo (2003) also observed Department of Natural Resources. Similar to the restored that flooding influences larger scale spatial patterns of wetlands, natural wetlands are depressional, precipitation- soil nutrients in floodplain forests, though vegetation also fed wetlands underlain by Adrian drained muck (sandy, influenced soil properties, particularly C and N, at smaller mixed, euic, mesic Terric Haplosaprists) and dominated by scales. Schlesinger et al. (1996) also found strong spatial Calamagrostis canadensis (Michx.) P. Beauv. and Scirpus patterns in soil N in desert soils, with spatial variation cyperinus (L.) Kunth. resulting from distributions of different shrubs and grasses. Conversely, Burke et al. (1999) showed that topography, not vegetation, influenced soil C pools in a shortgrass Soil sampling steppe, with C pools greater at toeslope positions. Collectively, these processes interact to lead to the creating We established a 32 × 32 m sampling grid in each wetland of microsite variability within the soil profile. in August 2010. Each grid was partitioned into four rows Often, however, comparisons of ecosystem services and four columns for a total of 16 subgrids in each site between natural and restored wetlands ignore their under- (Figure 2). The centroid of the each subgrid was marked lying spatial structure. Failing to take into account the with polyvinyl chloride pipe, and three randomly selected spatial variability of soil properties and processes increases soil cores (8.5 cm diameter × 5-cm deep) were collected the difficulty in identifying factors controlling ecosystem around each centroid for a total of 48 soil cores per wet- services such as C sequestration, nutrient accumulation, and land (192 cores total). Initially, 15-cm-deep cores were denitrification. We investigated the effects of restoration on taken and sectioned into 0- to 5-cm and 5- to 15-cm the spatial variability of denitrification in two restored and sections; however, only the 0- to 5-cm samples were two natural depressional wetlands. Our hypotheses were (1) analyzed to keep the sample size manageable. Cores natural wetlands would have greater variability of denitrifi- were collected with a stainless-steel piston corer, placed cation and associatedsoilpropertiesthanrestoredwetlands, into resealable plastic bags, and returned to the laboratory and (2) denitrification and soil properties would be spatially on ice. All soils were collected over 2 days in late August autocorrelated in natural wetlands, whereas no spatial after at least 1 week of no rain. Restored wetlands were 38 J.M. Marton et al. Newton County, Indiana, USA Legend Natural Wetlands Restored Wetlands G ^ 0 3 6 12 18 24 Kilometers Figure 1. Map of study sites in northwestern Indiana. Polygon represents property boundary of Kankakee Sands Nature Preserve. sampled on the first day and natural wetlands were in mass of a 5-g sample after drying to a constant weight. sampled on the following day. All soil samples were Plant-available N was extracted from soil samples using 2N processed within 48 hr of collection. Following the soil KCl (Mulvaney 1996) and analyzed using a Lachat collection, surface water was collected in 20-L carboys, Quickchem (Lachat Instruments, Loveland, CO, USA). returned to the laboratory, and filtered through 0.20-µm Unamended denitrification was measured with the acety- mesh for use in denitrification assays. lene-inhibition method (Tiedje 1994). Twenty-five grams of field-moist soil were placed into 125-mL Wheaton bottles with screw caps equipped with gray butyl septa. Each bottle Soil analyses received 50 mL of filtered site water amended with chlor- amphenicol (0.21 µM) to inhibit microbial growth. The Field-moist soils were analyzed for soil moisture content, bottles were flushed for 5 minutes with ultra-high-purity plant-available N (NO -N, NH -N), and denitrification. Soil 3 4 N gas, and the headspace was adjusted to 10% atm with moisture content was determined by measuring the change 2 International Journal of Biodiversity Science, Ecosystem Services & Management 39 least in one wetland. Data were tested for normality using Centroid the Kolmogorov–Smirnov test (α = 0.05) and, when neces- Sample sary, were natural-log-transformed to achieve a normal dis- tribution (IBM SPSS, Armonk, NY, USA). Means and standard errors were calculated for each soil property, and the variance between natural and restored wetlands was compared using Levene’stest (α = 0.01). Due to the weak observed autocorrelation, means were compared with a one- way ANOVA with Tukey’s post hoc test. Omnidirectional variograms were generated for denitri- fication and associated soil properties to evaluate their actual spatial structure in each of the four wetlands. Variograms describe how data are correlated with distance, typically in a specified direction; omnidirectional variograms essentially 32 m combine all directional variograms into a single plot, indi- Figure 2. Representative sampling scheme. Three samples were cating overall spatial continuity or structure in empirical collected at random directions and distances up to 4 m from each data. They indicate mean variability between multiple of the 16 centroids for a total of 48 soil cores per site. Lag values pairs of points across all lag distances in each possible ranged from 3.58 to 40.56 m with the number of neighbors direction (Eastman 2009). In order to test for the presence within a lag distance ranging from 2 to 30. of spatial autocorrelation among soil properties in each wet- land, global Moran’s I was calculated. Moran’s I tests for acetylene to block the reduction of N OtoN .Incubations autocorrelation among samples by comparing the variation 2 2 were conducted for 90 minutes at 25°C and a 5-mL sample of paired points within a specified lag distance (spatial was collected after 30, 60, and 90 minutes and stored in 2- covariation) to the total variance (Moran 1950; Legendre mL evacuated Wheaton vials sealed with aluminum crimp & Fortin 1989). The significance of the test is estimated by tops and gray butyl septa. The bottles were vigorously generating a z-score from a randomly permutated distribu- shaken by hand for 30 seconds prior to the collection to tion (Moran 1950; Legendre & Fortin 1989). Values range equilibrate N O between the soil slurry and headspace. from −1 indicating that the neighboring values are dissimilar After collection, a 5-mL mixture of N and acetylene (9:1) (negative spatial autocorrelation) to +1 indicating that neigh- was added to maintain constant pressure. Nitrous oxide boring values are similar (positive spatial autocorrelation). concentrations were measured using a gas chromatograph The Moran’s I test was performed over a range of lag (SRI Instruments, Menlo Park, CA, USA) with an electron distances from the lowest value at each site up to 20 m. capture detector, and concentrations were corrected for dilu- The lowest lag distance is the shortest distance between tion through multiple sample collections. Denitrification points such that each location has at least one neighbor. rates were determined by regressing N O concentrations The global Moran’s I test is a ‘global’ statistic of spatial against time. All denitrification rates were expressed on autocorrelation, meaning that clusters of high values (hot both a dry weight basis by correcting for the soil moisture spots) and low values (cold spots) are not identified (Moran content and an area basis by multiplying dry weight rates by 1950). The Getis–Ord Gi* test (Getis & Ord 1992) was used the sampling depth and bulk density. to detect hot spots of denitrification in each wetland. Hot Following analyses of field-moist soils, remaining spots (and cold spots) are identified as clusters with signifi- soils were dried, ground, and passed through a 2-mm cantly greater (or lower) values relative to surrounding mesh sieve to determine bulk density, organic C, and points. Omnidirectional variograms were generated using total N. Carbonates were removed by placing subsamples IDRISI Taiga (Clark Labs, Worcester, MA, USA), and all in a dessicator with a beaker of concentrated HCl for 24 hr other geostatistical analyses were performed using GeoDa (Hedges & Stern 1984). Organic C and total N were (Luc Anselin, University of Illinois, Urbana-Champaign, determined from these subsamples using a Perkin-Elmer Urbana, IL, USA; Anselin et al. 2006). 2400 CHN Analyzer (Perkin-Elmer, Waltham, MA, USA). Bulk density was calculated by dividing the total dry Results weight of the soil sample by the volume of the core (Blake & Hartge 1986). All results were expressed on a Denitrification was greater in restored wetlands (30 ng −1 −1 dry gram basis. NOg hr ) than in natural wetlands (1.1 ng −1 −1 N Og hr ) and was most strongly correlated to C:N ratios (rho = 0.50, p < 0.01). Due to the differences in bulk Geostatistics and data analysis density between the natural and restored wetlands, denitri- A priori, soil cores within each wetland were not considered fication was corrected for the specific bulk density and independent samples due to their close proximity to one sampling depth to express rates based on area. When another (Bruland et al. 2006). All variables exhibited weak expressed on a mass basis, the denitrification between the but significant positive spatial autocorrelation (see below) at two natural wetlands significantly differed (Figure 3a); 32 m 40 J.M. Marton et al. restored wetlands regardless of units. Because rates and A B B patters were comparable regardless of units, rates based on the dry weight were used for geostatistical analysis. Restored wetlands also had greater bulk density and C:N ratios, whereas natural wetlands had greater soil moisture, plant-available N (NO -N, NH -N), organic 3 4 C, and total N (Table 1). Levene’stest(α =0.01) indi- cated significantly greater variation in denitrification, soil moisture, bulk density, total N, and C:N ratios (Table 1) in restored wetlands than in natural wetlands. –2 Natural wetlands, on the other hand, had greater var- –4 iance of NO -N. Plant-available NH -N and organic C 3 4 had comparable variation between natural and restored 10 wetlands. B C A C Overall, omnidirectional variograms did not indicate marked spatial trends in denitrification or other measured variables across the individual natural or restored wetlands (data not shown). Global Moran’s I analyses revealed a wide range of spatial autocorrelation among soil properties and the four wetlands, at variable lag distances. Denitrification exhibited positive spatial autocorrelation in both natural wetlands between 4 and 5 m and up to 16 m in restored wetland 1; no spatial autocorrelation was –2 detected at any lag distance in restored wetland 2 (Table 2, –4 Figure 4). Organic C was autocorrelated in restored wet- 1 2 1 2 land 2 only up to 20 m, whereas no spatial autocorrelation was found in natural wetlands or restored wetland 1 (Table 2, Figure 5). Conversely, NO -N was spatially Figure 3. Boxplots of denitrification (natural-log transformed) 3 from the two natural and restored wetlands (n = 48). Different autocorrelated in both natural wetlands, but only in one letters indicate significant differences based on Tukey’s test. of the two restored wetlands (Table 2). The two natural Variance was significantly greater in restored wetlands based on wetlands exhibited similar patterns for both denitrification Levene’s test (p < 0.05). and soil organic C (Figures 4 and 5, respectively). However, denitrification in restored wetland 2 had a com- parable pattern to the natural wetlands (Figure 4), whereas however, no differences were detected between these two soil organic C in restored wetland 1 mimicked the natural sites when expressed on an area basis (Figure 3b). wetlands (Figure 5). Denitrification patterns were comparable between the two Table 1. Site means (±1 SE) of soil properties at each of the sampled wetlands. Moisture −3 −1 −1 (%) Bulk density (g cm )NO -N (µg g )NH -N (µg g ) Organic C (%) Total N (%) C:N ratio (mol:mol) 3 4 Restored 1 36 ± 1.7 0.83 ± 0.03 4.6 ± 0.9 0.86 ± 0.14 2.5 ± 0.31 0.15 ± 0.03 22.6 ± 1.2 Restored 2 11 ± 0.9 1.1 ± 0.03 3.8 ± 1.0 2.0 ± 0.35 1.9 ± 0.13 0.16 ± 0.01 14.2 ± 0.11 Natural 1 47 ± 1.2 0.35 ± 0.02 26 ± 2 5.9 ± 0.65 16 ± 1 1.5 ± 0.08 12.5 ± 0.07 Natural 2 46 ± 1.4 0.40 ± 0.02 79 ± 7 1.8 ± 0.20 16 ± 1 1.3 ± 0.08 13.7 ± 0.12 Mean a a b a a Restored 23 ± 1.6 0.97 ± 0.03 4.1 ± 0.7 1.4 ± 0.19 2.2 ± 0.17 0.15 ± 0.01 18.4 ± 0.75 b b a b b Natural 46 ± 1.0 0.37 ± 0.02 53 ± 4 3.8 ± 0.40 16 ± 1 1.4 ± 0.06 13.1 ± 0.09 Note: Different letters indicate significantly different variances between natural and restored wetlands based on Levene’s test (α = 0.01). Table 2. Lag distances (m) with significant spatial autocorrelation in the two natural and two restored depressional wetlands based on Moran’s I analysis. Denitrification Bulk density Soil moisture NO -N NH -N Organic C Total N C:N 3 4 Restored wetland 1 16 20 – – 5.8 –– – Restored wetland 2 – 7.6 20 12.1 – 20 20 9.1 Natural wetland 1 5.1 – 20 5.1 –– – – Natural wetland 2 4.2 –– 5.8 7.7 –– – Note: No significant spatial autocorrelation present at any lag distance. –2 –1 –1 –1 Denitrification (ng N O cm hr ) Denitrification (ng N O g hr ) 2 2 (LN-Transformed) (LN-Transformed) International Journal of Biodiversity Science, Ecosystem Services & Management 41 Denitrification Soil Organic C 0.4 0.5 Restored 1 Restored 1 Restored 2 0.4 Restored 2 Natural 1 Natural 1 0.3 Natural 2 0.3 Natural 2 0.2 0.2 0.1 0.1 0.0 –0.1 0.0 –0.2 –0.1 –0.3 2 4 6 8 10 12 14 16 18 20 2468 10 12 14 16 18 20 Lag (m) Lag (m) Figure 4. Moran’s I values for denitrification across lag dis- Figure 5. Moran’s I values for soil organic C across lag dis- tances in natural and restored wetlands. tances in natural and restored wetlands. Getis–Ord Gi* results indicated significant hot spots in natural wetland 2. Hot spots and cold spots were also of denitrification in all four wetlands (Figure 6). Restored identified for other measured soil properties, though wetland 1 had 6 significant hot spots and restored wet- denitrification hot spots did not correspond to those land 2 only had one. Six hot spots were detected in of NO -N, soil moisture, or organic C in any of the natural wetland 1 and only three hot spots were detected wetlands. (a) (b) Restored 2 Restored 1 Not Not Significant Significant Low Low High High 0 8 16 24 32 0 8 16 24 32 X Distance (m) X Distance (m) Natural 2 Natural 1 (c) (d) Not Not 32 32 Signficant Significant Low Low High High 24 24 0 0 0 8 16 24 32 0 8 16 24 32 X Distance (m) X Distance (m) Figure 6. Denitrification hot spots and cold spots in the (a) restored wetland 1, (b) restored wetland 2, (c) natural wetland 1, and (d) natural wetland 2 based on the Getis–Ord Gi* test (α = 0.05). Moran's I Y Distance (m) Y Distance (m) Y Distance (m) Y Distance (m) Moran's I 42 J.M. Marton et al. Discussion For most soil properties, restored wetlands had greater variance (Table 1) and greater significant lag distances Higher denitrification in restored wetlands relative to nat- (Table 2), though causal processes behind the observed ural wetlands was an unexpected result. Other studies have spatial trends are not entirely clear. Bruland and found greater denitrification in natural wetlands than in Richardson (2005) also measured significant spatial trends restored wetlands in multiple hydrogeomorphic settings. in created, restored, and natural riverine and non-riverine Hunter and Faulkner (2001) measured greater denitrification −1 −1 wetlands. They suggested that overbank flooding from the rates in natural bottomland wetlands (657 ng N Og hr ) adjacent river could have led to increasing soil organic compared to restored bottomland wetlands (167 ng −1 −1 matter and decreasing sand content with increasing dis- N Og hr ), and Bruland et al. (2006) found greater tance from the stream. Bruland and Richardson (2005) denitrification in natural riverine and non-riverine wetlands also found linear and non-linear trends in soil properties compared to created and restored wetlands. In both studies, from non-riverine wetlands, which they hypothesized to be positive correlations were detected between denitrification related to variations in unmeasured properties such as and soil moisture. Using a larger set of wetlands (n =20), topography, above- and belowground biomass, and soil including the four used in this study, Marton et al. (2014) biota (Schlesinger et al. 1996; Ettema & Wardle 2002). measured greater ambient and potential denitrification in These study sites were depressional systems, likely having natural wetlands than in restored wetlands, with denitrifica- minimal horizontal movement of surface water. In contrast tion rates positively correlated to soil NO -N and moisture. to our study, Bruland et al. (2006) found lower variation in In the current study, denitrification was positively correlated denitrification and NO -N in restored non-riverine wet- to C:N ratios (rho = 0.50, p < 0.01) and negatively corre- lands compared to natural wetlands. Conversely, they lated to both NO -N (rho = −0.50, p < 0.01) and soil found equal or greater variation of soil moisture and organic C (rho = −0.34, p < 0.01). One possible explanation NH -N in restored non-riverine wetlands. Their restored is that the N O produced during incubations in the current wetlands were former agricultural fields, and they reported study was the result of dissimilatory nitrate reduction to that surface soils were homogenized during restoration. ammonia (DNRA) rather than denitrification, which can However, their results were not consistent between river- occur in substrates with high C:N ratios and produces ine and non-riverine wetlands. In one of their riverine N O as a by-product (Burgin & Hamilton 2007). In created sites, spatial variability of denitrification was greater in freshwater wetlands in Texas, Scott et al. (2008)found the restored wetland than in the natural wetland. Orr et al. DNRA occurring simultaneously with denitrification, (2014) also found greater denitrification rates in a restored though DNRA frequently accounted for a small portion floodplain wetland relative to a natural floodplain wetland. (<5%) of total NO -N removal. However, they did find They found greater variability and no spatial structure in that DNRA accounted for up to 36% of NO -N removal denitrification at the restored site, consistent with our in one location characterized by low overlying water NO - findings. However, denitrification exhibited a strong spa- N concentrations. In a review of aquatic N-cycling pro- tial structure in their natural site. The lack of a significant cesses, Burgin and Hamilton (2007) discuss the variability spatial structure in the current study suggests that means of DNRA in total NO -N removal across freshwater and and variances of denitrification and soil properties across marine systems, and that this process may be less important the sampled natural and restored wetlands are not spatially in freshwater wetlands. The magnitude of DNRA in the dependent or vary at spatial scales greater or smaller than current study, while not known, could have implications for the scale at which these measurements were made. the overall delivery of ecosystem services. If, for example, Another possibility is the complexity of wetland eco- DNRA is a more important NO -N removal pathway than systems, with intra-site variability, soil and hydrological denitrification, then N O production may exceed N pro- 2 2 dynamics, and disturbance regimes make it difficult to duction, which is a less favorable outcome as N Ois a accurately parse out the spatial variability in these soil highly potent greenhouse gas. properties and processes. Several factors, alone or acting Soil organic C was greater in natural wetlands relative in concert, can influence the spatial distribution and struc- to restored wetlands. Development of soil organic matter is ture of soil properties, which in turn can influence the a slow process (Craft et al. 1999; Ballantine & Schneider spatial structure and magnitude of denitrification. 2009), and restored wetlands in the current study were only Following the end of cultivation, vegetation communities 10 years old. Further, the prairies surrounding the restored can go through successional stages, which in turn can wetlands are burned every 2–3 years. The high sand content influence soil properties such as C and N pools, soil of the restored wetlands (91% sand, Marton et al. 2014) moisture, oxygen content, and ultimately denitrification promotes drainage to groundwater, thereby allowing fires to (Gross et al. 1995; Rotkin-Ellman et al. 2004; Diekmann burn through the wetlands, which may inhibit the buildup et al. 2007). Bachand and Horne (1999) found greater of soil organic C (Neff et al. 2005). Anderson et al. (2005) rates of NO -N removal in a constructed wetland with a found that soil organic matter concentrations increased over mixed vegetation community of Typha spp. and Scirpus 10 years in two created riverine marshes in Ohio, while also spp. relative to individual monocultures. Gross et al. increasing in spatial variability due to variability in inunda- (1995) found greater spatial variability in soil N in mid- tion, vegetation, and sediment deposition. International Journal of Biodiversity Science, Ecosystem Services & Management 43 successional fields relative to a newly abandoned field and 2003). Important factors controlling denitrification rates a forest, while Rotkin-Ellman et al. (2004) showed that are NO -N, labile C, and anoxic conditions (Hunter & different tree species lead to differences in soil organic Faulkner 2001; McClain et al. 2003; Bruland et al. 2006; matter patches, which can be hot spots of denitrification. Ullah & Faulkner 2006; Bruland et al. 2009). Generally, Soil texture between the natural and restored wetlands in hot spots of denitrification and controlling factors should the current study was relatively comparable (Marton et al. correspond. In the current study, however, denitrification 2014), though there soil organic matter was greater in was not correlated with NO -N, soil organic C, or soil natural wetlands relative to restored wetlands. Though moisture. Further, denitrification hot spots did not overlap not statistically greater, the restored wetlands had greater with variables required for denitrification (e.g., soil moist- plot scale (1 m ) and site-level species richness relative to ure, NO -N, organic C), suggesting that other factors may the natural wetlands (Hopple & Craft 2013). The greater have been influencing denitrification. Plausible factors that plant diversity, coupled with differences in organic matter may explain denitrification include differences in micro- and slight variations in soil texture, could have led to the bial communities, soil texture, or interactions between observed differences in magnitude and spatial structure of NO -N, organic C, and moisture (Hanson et al. 1994; denitrification and other soil properties (e.g., soil organic Bruland et al. 2006; Peralta et al. 2010). Also, because C) between and within natural and restored wetlands. denitrification is a microbial process, our scale of measure- The observed spatial variance in soil properties from ment at the nearest tenth of a meter may have been too natural and restored wetlands was not entirely the result of coarse to detect relationships between denitrification and random processes. Significant positive spatial autocorrela- associated soil properties. tion for denitrification and several other soil properties Other studies have focused on identifying hot spots of (Figures 4 and 5, Table 2) suggested that, in addition to denitrification and soil properties, though hot spots were the plot-scale directional trends, significant spatial variabil- identified visually either based on interpolated maps or ity was present as well. Though the Moran’s I analysis based on parametric statistics that ignore spatial autocor- shows whether variables are globally autocorrelated over relation among samples (Rotkin-Ellman et al. 2004; space, the test does not provide insight into the causality of Bruland & Richardson 2005; Bruland et al. 2006). Our the spatial patterns. Future work could incorporate greater study differed in that we used the Getis–Ord Gi* test to sample sizes and attempt explanatory models, such as spa- statistically verify hot spots of denitrification. Ignoring the tially weighted regressions. Spatial patterns in vegetation, spatial aspect of soil properties and processes in attempt- hydrodynamics, and soil biota have been found to influence ing to estimate hot spots artificially inflates the signifi- soil nutrient pools and processes (Schlesinger et al. 1996; cance of the statistical tests by treating each sample as an Ettema & Wardle 2002;King et al. 2004). Schlesinger et al. independent observation, though due to their proximity in (1996) found that available N (NO -N, NH -N) was space, the samples cannot truly be considered independent 3 4 strongly autocorrelated within 20 cm of the perennial (Moran 1950; Legendre & Fortin 1989; Goovaerts 1998). bunchgrass Bouteloua eriopoda in desert soils, resulting The comparable, and in some cases greater, spatial from belowground nutrient cycling by the shrubs and asso- variability in denitrification and soil properties in restored ciated soil microorganisms. In a restored riparian wetland in wetlands relative to natural wetlands was surprising. Georgia, Ettema et al. (1998) found that soil properties Previous research has suggested that prior land use affects (NO -N, soil moisture) exhibited strong spatial autocorrela- soil properties, particularly agriculture that homogenizes tion. Plant-available NO -N was autocorrelated at distances soil structure both vertically and horizontally (Bruland up to 84 m, whereas in the current study, NO -N was et al. 2003). We found the opposite trend in that denitrifi- autocorrelated at distances up to 8 m. One possible expla- cation in restored wetlands exhibited stronger spatial varia- nation for the large discrepancy between distances in the bility and a comparable numbers of hot spots relative to two studies is the influence of hydrology. Ettema et al. natural wetlands. These results suggest that, despite prior (1998) measured soil variables in a riparian wetland that cultivation and homogenization of soil profiles, restoration received periodic pulses of water from the adjacent river, can result in systems with comparable ranges of biogeo- distributing sediment and nutrients over greater distances. chemical functioning. Conversely, our study sites were precipitation-fed depres- There are both advantages and disadvantages of our sional wetlands, which received little in the way of sampling design. Our sample plots were only 32 × 32 m allochthonous inputs. King et al. (2004)measured the var- which prevented us from adequately sampling the entire iation and relationships between environmental factors and wetland and thereby capturing site-scale variability. vegetation communities along a 10-km transect of wetlands However, we were able to sample four wetlands in total in the Everglades. They found that nutrients (K, N, and P) allowing us to make comparisons both between and within and hydrology influenced vegetation patterns at the wetland natural and restored depressional wetlands. For example, scale, whereas P loading controlled vegetation community though spatial patterns of denitrification and soil organic C patterns along the 10-km transect. were comparable between the two natural wetlands, the Denitrification hot spots occur where reactants and patterns differed between the two restored sites. Though it conditions suitable for denitrification are disproportio- cannot be determined why these differences exist, it is nately higher than the surrounding areas (McClain et al. possible that differences in plant density and diversity 44 J.M. Marton et al. between the two restored sites are influencing below- of Agriculture, Natural Resources Conservation Service, Conservation Effects Assessment Program through The Great ground dynamics. Lakes-Northern Forest Cooperative Ecosystem Studies Unit, With a smaller extent and finer grain, we were able to Cooperative Agreement Number 68-7482-9-516. investigate patterns in denitrification and the associated soil properties over shorter spatial scales. Changing either the extent or the grain of the study area would ultimately change the variability, and thus the predictability, of a References given parameter (Wiens 1989). This combined with the Anderson C, Mitsch W, Nairn R. 2005. Temporal and spatial fact that we were unable to fit a reliable semivariogram to development of surface soil conditions at two created river- any of our variables implies an inherent difficulty and ine marshes. J Environ Qual. 34:2072–2081. Anselin L, Syabri I, Kho Y. 2006. GeoDa: an introduction to potential inapplicability in scaling up our results to the spatial data analysis. Geogr Anal. 38:5–22. wetland scale. Though these results may not be effectively Bachand P, Horne A. 1999. Denitrification in constructed free- scaled to the wetland scale, findings from this and other water surface wetlands: II. Effects of vegetation and tem- similar studies (Bruland & Richardson 2004, 2005; Orr perature. Ecol Eng. 14:17–32. et al. 2014) can provide useful information guiding wet- Ballantine K, Schneider R. 2009. Fifty-five years of soil devel- opment in restored freshwater depressional wetlands. Ecol land restoration and post-restoration monitoring and Appl. 19:1467–1480. assessment. For example, reestablishment of hydrology, Blake G, Hartge K. 1986. Bulk density. In: Klute A, editor. native vegetation, and organic matter amendments will Methods of soil analysis, part 1. Physical and mineralogical help a restored system have comparable or greater deni- methods: agronomy monograph no. 9. 2nd ed. Madison trification potential when compared to natural reference (WI): American Society of Agronomy; p. 363–375. Bruland G, Hanchey M, Richardson C. 2003. Effects of agricul- sites. Further, these results also show that a high number ture and wetland restoration on hydrology, soils, and water of samples can provide a good estimation of the variability quality of a Carolina bay complex. Wetl Ecol Manage. of soil properties and processes, though the sampling 11:141–156. extent would have to cover the majority of the wetland Bruland G, Richardson C. 2004. A spatially explicit investigation to fully understand the spatial structure and to appropri- of phosphorus sorption and related soil properties in two riparian wetlands. J Environ Qual. 33:785–794. ately scale those findings. While it is not feasible to con- Bruland G, Richardson C. 2005. Spatial variability of soil proper- duct spatially intensive investigations of denitrification in ties in created, restored, and paired natural wetlands. Soil Sci most cases, more studies that are conducted with a spatial Soc Am J. 69:273–284. sampling design will increase our understanding of the Bruland G, Richardson C, Daniels W. 2009. Microbial and controls on spatial structuring of ecosystem services and geochemical responses to organic matter amendments in a created wetland. Wetlands. 29:1153–1165. how this structure changes over time. Bruland G, Richardson C, Whalen S. 2006. Spatial variability of Incorporating spatial patterns provides greater under- denitrification potential and related soil properties in created, standing of how individual and multiple environmental restored, and paired natural wetlands. Wetlands. 26:1042–1056. factors influence the distribution and development of soil Burgin A, Hamilton S. 2007. Have we overemphasized the role properties and processes. This is particularly true for of denitrification in aquatic ecosystems? A review of nitrate removal pathways. Front Ecol the Environ. 5:89–96. restoring wetlands, in which the primary goal is to return Burke I, Lauenroth W, Riggle R, Brannen P, Madigan B, Beard some ecosystem service to the landscape. Our results S. 1999. Spatial variability of soil properties in the showed that, despite lower concentrations of soil C and Shortgrass Steppe: the relative importance of topography, N, restored wetlands had more complex spatial variability grazing, microsite, and plant species in controlling spatial in denitrification, soil moisture, bulk density, total N, and patterns. Ecosystems. 2:422–438. Craft C, Reader J, Sacco J, Broome S. 1999. Twenty-five years C:N ratios compared to natural wetlands and provided a of ecosystem development of constructed Spartina alterni- greater range of biogeochemical transformations. flora (Loisel) marshes. Ecol Appl. 9:1405–1419. Increased knowledge of the spatial structure of soil proper- Diekmann L, Lawrence D, Okin G. 2007. Changes in the spatial ties will further reveal factors controlling processes such variation of soil properties following shifting cultivation in a as denitrification at multiple spatial scales in restored wet- Mexican tropical dry forest. Biogeochemistry. 84:99–113. Eastman J. 2009. IDRISI Taiga: guide to GIS and image proces- lands. By understanding the underlying factors driving sing. Worcester (MA): Clark Labs, Clark University. denitrification and other soil processes (e.g., accretion, Ettema C, Coleman D, Vellidis G, Lowrance R, Rathbun S. 1998. sedimentation, P sorption), wetland restoration can be Spatiotemporal distributions of bacterivorous nematodes and improved to maximize the delivery of ecosystem services. soil resources in a restored riparian wetland. Ecology. 79:2721–2734. Ettema C, Wardle D. 2002. Spatial soil ecology. Trends Ecol Evol. 17:177–183. Acknowledgments Fennessy S, Craft C. 2011. Agricultural conservation practices We would like to thank Chip O’Leary and Stephanie Frischie at increase wetland ecosystem services in the Glaciated Interior The Nature Conservancy at the Kankakee Sands Preserve and the Plains. Ecol Appl. 21:S49–S64. Indiana Department of Natural Resources for access to their Gallardo A. 2003. Spatial variability of soil properties in a flood- property. We would also like to thank Ellen Herbert, Anya plain forest in Northwest Spain. Ecosystems. 6:564–576. Hopple, and Bri Richards for help with the collection, prepara- Getis A, Ord J. 1992. The analysis of spatial association by use tion, and analysis of samples, and Maggie Marton for editorial of distance statistics. Geogr Anal. 24:189–206. assistance. This study was funded by the US Department International Journal of Biodiversity Science, Ecosystem Services & Management 45 strategies to counter a persistent ecological problem. Gittins R. 1968. Trend-surface analysis of ecological data. J Ecol. BioScience. 51:373–388. 56:845–869. Moran P. 1950. Notes on continuous stochastic phenomena. Goovaerts P. 1998. Geostatistical tools for characterizing the Biometrika. 37:17–23. spatial variability of microbiological and physico-chemical Mulvaney R. 1996. Nitrogen- inorganic forms. In: Sparks DL, soil properties. Biol Fertil Soils. 27:315–334. editor. Methods of soil analysis, part 3: chemical methods. Gross K, Pregitzer K, Burton A. 1995. Spatial variation in nitro- Madison (WI): Soil Science Society of America; p. 1123– gen availability in three successional plant communities. J Ecol. 83:357–367. Neff J, Harden J, Gleixner G. 2005. Fire effects on soil organic Grunwald S, Reddy K, Newman S, DeBusk W. 2004. Spatial matter content, composition, and nutrients in boreal interior variability, distribution and uncertainty assessment of soil Alaska. Can J For Res. 35:2178–2187. phosphorus in a south Florida wetland. Environmetrics. Orr C, Stanley E, Wilson K, Finlay J. 2007. Effects of restoration 15:811–825. and reflooding on soil denitrification in a leveed Midwestern Grunwald S, Reddy K, Prenger J, Fisher M. 2007. Modeling of floodplain. Ecol Appl. 17:2365–2376. the spatial variability of biogeochemical soil properties in a freshwater ecosystem. Ecol Modell. 201:521–535. Orr CH, Predick KI, Stanley EH, Rogers KL. 2014. Spatial Hanson G, Groffman P, Gold A. 1994. Denitrification in riparian autocorrelation of denitrification in a restored and a natural wetlands receiving high and low groundwater nitrate inputs. floodplain. Wetlands. 34:89–100. J Environ Qual. 23:917–922. Osborne L, Kovacic D. 1993. Riparian vegetated buffer strips in Harms TK, Wentz EA, Grimm NB. 2009. Spatial heterogeneity water-quality restoration and stream management. Freshw of denitrification in semi-arid floodplains. Ecosystems. Biol. 29:243–258. 12:129–143. Peralta A, Matthews J, Kent A. 2010. Microbial community Hedges J, Stern J. 1984. Carbon and nitrogen determinations of structure and denitrification in a wetland mitigation bank. carbonate-containing solids. Limnol Oceanogr. 29:657– Appl Environ Microbiol. 76:4207–4215. 663. Rabalais NN, Turner RE, Scavia D. 2002. Beyond science into policy: Gulf of Mexico hypoxia and the Mississippi River. Hopple A, Craft C. 2013. Managed disturbance enhances biodi- BioScience. 52:129–142. versity of restored wetlands in the agricultural Midwest. Ecol Rotkin-Ellman M, Addy K, Gold A, Groffman P. 2004. Tree Eng. 61:505–510. species, root decomposition and subsurface denitrification Hunter R, Faulkner S. 2001. Denitrification potentials in restored potential in riparian wetlands. Plant Soil. 263:335–344. and natural bottomland hardwood wetlands. Soil Sci Soc Am Schlesinger W, Raikes J, Hartley A, Cross A. 1996. On the J. 65:1865–1872. spatial pattern of soil nutrients in desert ecosystems. King R, Richardson C, Urban D, Romanowicz E. 2004. Spatial Ecology. 77:364–374. dependency of vegetation-environment linkages in an anthro- Scott JT, McCarthy MJ, Gardner WS, Doyle RD. 2008. pogenically influenced wetland ecosystem. Ecosystems. Denitrification, dissimilatory nitrate reduction to ammonium, 7:75–97. and nitrogen fixation along a nitrate concentration gradient in Kovacic DA, Twait RM, Wallace MP, Bowling JM. 2006. Use of a created freshwater wetland. Biogeochemistry. 87:99–111. created wetlands to improve water quality in the Midwest- Tiedje J. 1994. Denitrifier enzyme activity. In: Weaver RW, Angle Lake Bloomington case study. Ecol Eng. 28:258–270. S, Bottomley P, editors. Methods of soil analysis, part 2: Legendre P, Fortin M. 1989. Spatial pattern and ecological ana- microbiological and biochemical properties. Madison (WI): lysis. Vegetatio. 80:107–138. Soil Science Society of America; p. 256–257. Marton JM, Fennessy MS, Craft CC. 2014. Functional differ- Ullah S, Faulkner S. 2006. Denitrification potential of different ences between natural and restored wetlands in the Glaciated Interior Plains. J Environ Qual. 43:409–417. land-use types in an agricultural watershed, lower McClain M, Boyer E, Dent C, Gergel S, Grimm N, Groffman P, Mississippi valley. Ecol Eng. 28:131–140. Hart S, Harvey J, Johnston C, Mayorga E, et al. 2003. Wiens J. 1989. Spatial scaling in ecology. Funct Ecol. 3:385–397. Biogeochemical hot spots and hot moments at the interface Woltemade C. 2000. Ability of restored wetlands to reduce of terrestrial and aquatic ecosystems. Ecosystems. 6:301–312. nitrogen and phosphorus concentrations in agricultural drai- Mitsch W, Day J, Wendell Gilliam J, Groffman P, Hey D, nage water. J Soil Water Conserv. 55:303–309. Randall G, Wang N. 2001. Reducing nitrogen loading to Zedler JB. 2003. Wetlands at your service: reducing impacts of the Gulf of Mexico from the Mississippi river basin: agriculture at the watershed scale. Front Ecol Environ. 1:65–72.

Journal

International Journal of Biodiversity Science, Ecosystem Services & ManagementTaylor & Francis

Published: Jan 2, 2015

Keywords: wetland restoration; denitrification; spatial variability; soil organic carbon; spatial autocorrelation

There are no references for this article.