| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Louisiana State University, Baton Rouge, LA 70803
* Corresponding author (jrwhite{at}lsu.edu).
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
6% of the total soil P released to the water column from the P enriched vegetated treatment compared with a 1% of total P released from the P enriched non-vegetated treatment. Initial P release rates from the enriched, vegetated treatment were five times greater than the enriched, non-vegetated treatment. Episodic growth of flood intolerant plants under drawdown conditions was shown to be a significant mechanism for nutrient release in ephemerally flooded P enriched wetland systems. Episodic flooding and drying cycles could therefore mobilize P over the long-term from P enriched to P unenriched areas. |
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Natural episodic drops in water levels (droughts) can lead to oxidation of wetland soils, increasing soil organic matter decomposition rates (Reddy, 1983; Fabre, 1988; Martin et al., 1996; Olila et al., 1997; Watts, 2000b; James et al., 2001; White and Reddy, 2001; DeBusk and Reddy, 2003). Organic soils therefore have the potential to release P to the water column on reflooding. Also, P retention capacity of soils that have undergone drawdown have been shown to diminish on reflooding compared with continually flooded soils (De Groot and Van Wijck, 1993; Qiu and McComb, 1994; Baldwin, 1996; Mitchell and Baldwin, 1998; Watts, 2000a; Klotz and Linn, 2001). However, little research has focused on the effect of initial soil P concentrations on water quality on reflooding in wetlands outside of the heavily studied Florida Everglades ecosystems (Newman and Pietro, 2001; Corstanje and Reddy, 2004).
Many P flux studies on soils or sediments have been conducted on intact soil cores devoid of vegetation (Holdren and Armstrong, 1980; Olila et al., 1997; Moore et al., 1998; Fisher and Reddy, 2001). However, vegetation is a key component in wetland ecosystems and community shifts are often seen as a result of water level fluctuations and soil nutrient concentrations (Yarbro, 1983; Gerritsen and Greening, 1989; Urban et al., 1993; Wu et al., 1997). Research has shown high soil P concentrations can result in vegetation shifts, leading to a concomitant loss of indigenous species (Davis, 1991, 1994; Föllmi, 1996). This vegetation can act as short-term P storage, which can rapidly release 35 to 75% of the total plant-associated P during senescence, potentially increasing water column P concentrations (Richardson, 1985; White et al., 2004; 2006; Corstanje et al., 2006).
The hypothesis of this study is that initial soil P (internal P load) and the presence of flood intolerant vegetation will increase P release to the water column after a drawdown/reflooding event. Specific objectives of this study were to: (i) compare P release rates from P enriched and unenriched soils, (ii) determine the P release rates for cores containing soil only and soil colonized with mature standing vegetation, and (iii) compare various pools of soil P.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
40 yr, surface inflows to the system were terminated in the early 1990s as part of a St. Johns River Water Management District restoration project. Previous studies documented the existence of a P gradient starting from the NE corner decreasing toward the interior of the marsh (Fig. 2 ) (Olila and Reddy, 1995). Higher soil total P (TP) levels (618 mg kg–1) were found closest to the inflow point of agricultural runoff along the NE levee. The Northwest (NW) region of the marsh has remained relatively unaffected by past surface water inputs with an average soil TP of 444 mg kg–1 (0–10 cm soil interval). Predominant and native vegetative species within the marsh are Cladium sp. stands and Panicum sp. Flats, now located primarily in the unenriched region. Typha latifolia encroached on Cladium sp. in the enriched region of the marsh, similar to trends documented in the Florida Everglades (Davis, 1991, 1994).
|
Following the simulated drought, the intact soil sections (30 cm i.d.) were subsampled using eight smaller cores (7 cm id) with four containing mature standing vegetation (vegetated, Eupatorium sp.) and four having no visible emergent vegetation (non-vegetated) for each of the P enriched and unenriched sites. Vegetated cores were collected by isolating individual plants in the center of each core. The core was then extracted by using a serrated knife and advancing the core tube down to minimize soil compaction. Both vegetated and non-vegetated cores contained belowground biomass and no roots were removed to minimize soil disturbance and to mimic field conditions. An additional three cores from each site were taken and sectioned to determine soil characteristics before reflood. All cores were then reflooded with filtered (0.45 µm) site water to produce an overlying water column of 30 cm. Water column P concentrations were corrected for repeated sampling and replacement of column water by filtered (0.45 µm) site water. Overlying water columns were maintained to a depth of 30 cm to prevent the overestimation of P concentrations due to increasingly reduced water column volume (Moore et al., 1998; DeBusk and Reddy, 2003; Malecki et al., 2004).
Site water P concentrations were not found to differ significantly between the two sites, with 0.012 and 0.025 mg L–1 for soluble reactive P (SRP) and TP, respectively. The reflooded cores were placed in a water bath with a mean temperature of 23°C. At termination of the experiment, there was insufficient plant biomass remaining in the vegetated treatment for determination of nutrient content due to decomposition processes, 38 d after the reflood.
Laboratory Analyses
Water Column
Water samples (40 mL) were taken from the mid-depth of the water column of each core on Days 0, 1, 2, 5, 10, 20, 30, and 38 after reflooding and analyzed for SRP. Total P was determined on all water samples except those taken on Days 20 and 30. Soluble reactive P samples were filtered through a 0.45-µm membrane filters and immediately frozen. The TP water samples were digested with 5.5 M (11 N) H2SO4 before colorimetric analysis (Method 365.1, USEPA, 1993). Filtered (0.45 µm) site water of known P concentration was added to the cores, equal to the amount removed at each sampling (40 mL), to maintain a consistent water column of 30 cm over the incubation period (Moore et al., 1998). The removal and addition of site water was accounted for in the calculation of the P flux rates.
Soil
At the end of the 38-d flux study, the top 0 to 10 cm of each soil core was removed and analyzed for the following physicochemical properties: moisture content as a percentage of total wet soil mass (constant weight, 70°C); dry-weight bulk density and pH (1:1 ratio). Total N (TN) and total C (TC) were determined on dried ground subsamples using a Carlo-Erba NA-1500 CNS Analyzer (Haak-Buchler Instruments, Saddlebrook, NJ) and TP was determined using the ignition method followed by acid digestion (Anderson, 1976). Total inorganic P (TPi) was measured on a 1 M HCl extraction (Reddy et al., 1998) with total organic P (TPo) calculated as the difference between TP and TPi.
Microbial biomass P (MBP) was calculated as the difference in TP between fumigated (CHCl3) and non-fumigated samples following a 0.5 M NaHCO3 extraction (Allen et al., 1974). The 0.5 M NaHCO3 reagent extracted the organic and inorganic labile P fractions including the microbial biomass and represented the total labile organic P pool. The residual fumigated samples were sequentially extracted following the P fractionation scheme developed by Ivanoff et al. (1998) to determine moderately labile, non-labile, and residual organic P pools. An inorganic P fractionation scheme determined inorganic P forms representing the readily exchangeable (1.0 M KCl), Fe/Al bound (0.1 M NaOH) and Ca/Mg bound P (0.5 M HCl) extractable pools (Reddy et al., 1998).
Data Analysis
Phosphorus flux rates from intact cores were calculated by regressing concentration over two time intervals (1–10 and 10–38 d) due to the biphasic nature of the concentration vs. time curves. Water column and soil characteristics were contrasted by analysis of variance to determine the effect of soil P concentrations and vegetation presence on soil P forms. Tukey-Kramer adjustment was used for the multiple comparison of means (
= 0.05) among all the groups. Water column and soil characteristics were contrasted by ANOVA to determine the effect of soil P concentrations and vegetation presence on soil P forms. Tukey-Kramer adjustment was used for the multiple comparison of means ( = 0.05) among all the groups.
The overall experimental design consisted of a fully randomized two way ANOVA, with a site effect and plant effect, resulting in the general form:
 | [1] |
i equals effect due to the i-th site effect, and assumed to be normally distributed with mean zero and standard deviation 
; ßj equals effect due to the j-th plant effect, and assumed to be normally distributed with mean zero and standard deviation ; 
ijk equals residual effect, assumed to be normally distributed with mean zero and standard deviation . The model generates treatment effect estimates for site (df = 1), plant (df = 1), and the plant x site interaction (df = 1), with LSD's for the pair wise comparisons. Residual Maximum Likelihood (REML) mixed effects models were used in all analyses (MIXED procedure in SAS, Version 9.0, 2003, SAS Institute, Inc., Cary, NC).
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
All inorganic P soil fractions (readily available P, Fe/Al bound P, or Ca/Mg bound P) were not significantly different between the P enriched and unenriched soils, nor vegetated and non-vegetated (Table 3). The differences in TP between the soils therefore, can be assumed to be in the organic fractions. There was no significant difference in the microbial biomass vs. the vegetation and non-vegetated treatments due to large variability, which was likely due to the fact that non-vegetated cores also contained roots from surrounding vegetated areas. As mentioned in the methods, we did not remove any roots to both prevent disturbance and to maintain the experiment close to field conditions where roots would be present under bare soil. In this fashion, any difference in P flux would be primarily a result of release from the aboveground biomass.
|
Soluble Reactive Phosphorus Release
Soluble reactive P accounted for the majority of P released to the water column (Fig. 3
). In the non-vegetative treatment, this was equivalent to 98.5 and 100% of TP for the P enriched and unenriched non-vegetated treatment. In the vegetative treatment, SRP concentrations equated to 65 and 98.5% for the P enriched vegetated and P unenriched soils, respectively.
|
A significant increase in SRP was seen over the first 10 d after reflood followed by a period of steady concentrations between Day 10 and Day 38. The data were therefore divided into two phases. An initial, first release phase (1–10 d) calculated the initial P release rates directly after reflood. The second phase release rates (10–38 d) were determined for water samples taken after an extended period of reflood. There was an initial spike in SRP concentration between Day 0 and Day 1 in all the cores, presumably as a result of soil suspension during reflooding, which occurs on rehydration of the dry, organic soils. The SRP concentrations stabilized after 24 h and therefore the release rates were calculated from Day 1.
Initial SRP release rates were significantly lower from the P unenriched non-vegetated treatment, at 9.25 ± 1.80 mg SRP m–2 d–1. These rates were in comparison with the higher release rate from the P enriched non-vegetated treatment (13.4 ± 4.88 mg SRP m–2 d–1) (Table 4). The second phase (10- 38 d) of SRP release was lower with rates of 7.05 ± 2.30 and 12.7 ± 4.57 mg m–2 d–1 for the unenriched, non-vegetated and enriched, non-vegetated soils, respectively. This differential release led to final SRP concentrations in the water column in the P enriched soil only treatment to be double the concentration than the P unenriched soil only treatment (Fig. 3A and 3B).
|
Fluctuations in SRP concentrations were observed in the P enriched, vegetated treatment, beginning on Day 5 and with a decrease in SRP concentrations from Day 10 to Day 30. The SRP concentration curve took on an S-shape in the vegetated cores only, while the non-vegetated treatment release remained linear (Fig. 3A). Since this oscillation in P concentration was only seen in the vegetated treatment, it is possible this artifact may be a result of epiphytic microorganisms present on the stems and leaves of plants (Preece and Dickinson, 1971; Collins, 1976; Blakeman, 1981; Morris et al., 1996).
Water column SRP concentrations in the P-unenriched, vegetated treatment were significantly lower than those observed for non-vegetated cores, for four of the eight sampling events (Fig. 3B). The first phase (1–10 d) release rates in the vegetated cores averaged 5.74 ± 3.60 mg SRP m–2 d–1. The non-vegetated treatment was not significantly different at 9.25 ± 1.80 mg SRP m–2 d–1 (Table 4). Therefore, the growth of vegetation in P-unenriched soils did not lead to an increase in SRP release to the water column on reflooding so plant colonization alone was not a single factor in P release, but rather the interaction of plant colonization and P enriched soils led to this mobilization (Fig. 3B and 3D).
These results have important ramifications for wetlands with regions of P enriched soils. In the P enriched soil, SRP was mobilized by the plants on reflooding released into the water column. Conversely, the vegetated, unenriched soil released no significant amounts of SRP on reflood. Therefore, plant colonization during dry periods in P enriched soils is a significant mechanism for P release from the soil and this process could result in the remobilization and redistribution of soil P from the P enriched to P unenriched sections of the wetland when water flow is restored.
Total Phosphorus Release
Final TP concentrations in the water column were significantly lower at Day 38 for the P unenriched soil at 1.40 ± 0.55 mg P L–1 than for the P enriched soil at 2.06 ± 0.80 mg P L–1 (Fig. 3C and 3D). First phase TP release rates were not significantly different in the P enriched, non-vegetated treatment (8.69 ± 4.80 mg m–2 d–1) versus the P unenriched, non-vegetated treatment (9.16 ± 1.71 mg m–2 d–1). Water column TP concentrations rapidly increased over the first 10 d to 2.47 mg L–1 in the P enriched, vegetated treatment, with four times more TP released (26.6 ± 18.0 mg TP m–2 d–1) compared with the P enriched, non- vegetated treatment (8.69 ± 4.80 mg TP m–2 d–1) (Fig. 3C, Table 4).
Total P release from the P enriched, non-vegetated treatment cores continued to increase after Day 10 with a second phase release rate of 12.7 ± 4.13 mg TP m–2 d–1, demonstrating a continual, consistent release of P over the 38 d incubation (Fig. 3C). Cores from P enriched, vegetated treatments showed a significant drop in P release rates to 0.891 ± 7.51 mg m–2 d–1 10 d post reflood. This decline in release rates suggests the rapid release of P may have reduced the P concentration gradient between the water column and soil, thereby reducing flux out of the soil. The presence of vegetation significantly increased the initial release of TP in the P enriched soils only. This P flux is likely a result of P leaching from the plant, similar to the observed higher release rates of SRP from the vegetated, P enriched soils and includes some dissolved and particulate P fractions which comprise the difference between the final water column concentrations of SRP and TP concentrations (Fig. 3A and 3C).
Water column TP concentrations in the non-vegetated and vegetated, unenriched treatments, were not significantly different (Fig. 3D). The first phase (1–10 d) TP release rate in the vegetated cores were 9.40 ± 3.40 mg m–2 d–1 while the non-vegetated cores released TP at a rate of 9.16 ± 1.71 mg m–2 d–1. This finding further demonstrates that there was no significant effect of vegetation on release of TP, as was seen for SRP in P unenriched soils. After the initial phase, the vegetated and non-vegetated P unenriched treatments followed similar trends with 10 to 38 d release rates of 8.00 ± 3.85 and 6.99 ± 2.56 mg m–2 d–1, respectively (Table 4).
While nutrient enriched wetland soils show a propensity for P release following a drawdown/reflood event, the unenriched wetland soils did not. These results also suggest that flood intolerant plant colonization and growth during periods of low water/drought do not lead to mobilization of soil P in all cases. Where soil P is elevated due to historic external loading, the presence of flood intolerant vegetation can increase P release on reflooding. For similar wetland peat soils, we have observed significant P release over the span of several weeks for repeated drydown event (Corstanje and Reddy, 2004; White te al., 2004, 2006). Consequently, continual fluctuations in water levels could ultimately drive the redistribution of the soil P (internal load) in nutrient impacted wetlands.
The release rates determined in this lab study are somewhat limited by the fact that a non-flow through system was used. Under field conditions, it is assumed that water flowing through the marsh would carry away P, maintaining a higher soil to water column gradient. The result of this is that our release rates are likely an underestimation of the amount of P released.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Since the termination of external nutrient inputs to BCMCA, internal nutrient dynamics have become an important regulator of the system's recovery. In this study we have shown that a drawdown and subsequent reflood can result in significant fluxes of P from the nutrient enriched soils, rates that can be significantly increased by the presence of vegetation. Regarding the overall distribution of P, release from P enriched soils would have the effect of decreasing the soil P concentrations in enriched areas while increasing the soil P concentrations in others (unenriched), eventually elevating the overall baseline soil P concentrations of the entire system. These results have serious consequences for any marsh restoration if the restoration goal is to attain pre-impact soil P concentrations or to prevent further eutrophication of the marsh. These goals might not be attainable without the physical removal of the P enriched soil from the system to prevent the continual spread of P into pristine marsh area.
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received for publication March 30, 2006.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. M. Malecki-Brown and J. R. White Effect of Aluminum-Containing Amendments on Phosphorus Sequestration of Wastewater Treatment Wetland Soil Soil Sci. Soc. Am. J., May 1, 2009; 73(3): 852 - 861. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Belmont, J. R. White, and K. R. Reddy Phosphorus Sorption and Potential Phosphorus Storage in Sediments of Lake Istokpoga and the Upper Chain of Lakes, Florida, USA J. Environ. Qual., March 25, 2009; 38(3): 987 - 996. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. White, L. M. Gardner, M. Sees, and R. Corstanje The Short-Term Effects of Prescribed Burning on Biomass Removal and the Release of Nitrogen and Phosphorus in a Treatment Wetland J. Environ. Qual., October 23, 2008; 37(6): 2386 - 2391. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Malecki-Brown, J. R. White, and K. R. Reddy Soil Biogeochemical Characteristics Influenced by Alum Application in a Municipal Wastewater Treatment Wetland J. Environ. Qual., October 24, 2007; 36(6): 1904 - 1913. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
