Published in J. Environ. Qual. 33:1638-1646 (2004).
© ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA
TECHNICAL REPORTS
Bioremediation and Biodegradation
Accumulation of Perchlorate in Aquatic and Terrestrial Plants at a Field Scale
Kui Tana,
Todd A. Andersonb,
Matthew W. Jonesa,
Philip N. Smithb and
W. Andrew Jacksona,*
a Department of Civil Engineering, Texas Tech University, Lubbock, TX 79410
b The Institute of Environmental and Human Health, Texas Tech University, Lubbock, TX 79409-1163
* Corresponding author (Andrew.jackson{at}coe.ttu.edu). 677 S. Segoe Rd., Madison, WI 53711 USA
Received for publication November 5, 2003.
 |
ABSTRACT
|
|---|
Previous laboratory-scale studies have documented perchlorate 
uptake by different plant species, but less information is available at field scale, where ClO–4 uptake may be affected by environmental conditions, such as distance to streams or shallow water tables, exposure duration, and species. This study examined uptake of ClO–4 in smartweed (Polygonum spp.) and watercress (Nasturtium spp.) as well as more than forty trees, including ash (Fraxinus greggii A. Gray), chinaberry (Melia azedarach L.), elm (Ulmus parvifolia Jacq.), willow (Salix nigra Marshall), mulberry [Broussonetia papyrifera (L.) Vent.], and hackberry (Celtis laevigata Willd.) from multiple streams surrounding a perchlorate-contaminated site. Results indicate a large potential for ClO–4 accumulation in aquatic and terrestrial plants, with ClO–4 concentration in plant tissues approximately 100 times higher than that in bulk water. Perchlorate accumulation in leaves of terrestrial plants was also dependent on species, with hackberry, willow, and elm having a strong potential to accumulate ClO–4. Generally, trees located closer to the stream had a higher ClO–4 accumulation than trees located farther away from the stream. Seasonal leaf sampling of terrestrial plants indicated that ClO–4 accumulation also was affected by exposure duration, with highest accumulation observed in the late growing cycle, although leaf concentrations for a given tree were highly variable. Perchlorate may be re-released into the environment via leaching and rainfall as indicated by lower perchlorate concentrations in collected leaf litter. Information obtained from this study will be helpful to understand the fate of ClO–4 in macrophytes and natural systems.
Abbreviations: BCF, bioconcentration factor • NWIRP, Naval Weapons Industrial Reserve Plant
|
INTRODUCTION
|
|---|
PERCHLORATE has been identified as a contaminant in soil, sediment, ground water, and surface water at numerous sites associated with perchlorate manufacture or usage, such as defense, aerospace, and several chemical industries (Damian and Pontius, 1999). In the United States, there are more than 150 facilities or users linked with perchlorate in 44 states (Motzer, 2001). Perchlorate affects thyroid function of humans and animals (Urbansky, 1998; Motzer, 2001). The USEPA has begun to establish a drinking water standard for perchlorate (Na et al., 2002).
Bioremediation appears to be one of the most applicable technologies to remove perchlorate (Urbansky, 1998; Logan, 2001; Nerenberg et al., 2002; Hunter, 2002). In situ treatment techniques for perchlorate include permeable biobarriers with vegetable oil and different forms of biostimulation with organic substrate (Hunter, 2002). Numerous ex situ techniques are currently in practice or under review for perchlorate remediation, including ion exchange, reverse osmosis, electrodialysis, iron-preloaded granular carbon adsorption, and bioreactors (Motzer, 2001; Na et al., 2002; Logan, 2001). However, in most cases, contaminated ground water and soil are difficult to remediate. The search for a cost-effective and an environmentally friendly treatment option is still ongoing, especially for low-level perchlorate contamination of ground water and surface water. The potential to remove perchlorate by phytoremediation has been reported, and three major mechanisms were found to be linked with phytoremediation of perchlorate: phytoaccumulation, phytodegradation, and rhizodegradation (Nzengung et al., 1999; Aken and Schnoor, 2002).
There is a preponderance of laboratory data on perchlorate uptake in plants. Parrot-feather [Myriophyllum aquaticum (Vell.) Verdc.] has been used successfully to remediate TNT-contaminated soils, and was recently proven to hyperaccumulate perchlorate, up to 1200 mg kg–1 fresh wt. (on the basis of fresh wet weight) in plant tissues (Susarla et al., 1999). Susarla et al. (2000) demonstrated that perchlorate can be taken up and transformed by several vascular plants, including smartweed (Polygonum punctatum Elliott), pickleweed [Allenrolfea occidentalis (S. Wats.) Kuntze], sweet gum (Liquidambar styraciflua L.), water lily (Nymphaea odorata Aiton), and black willow. Perchlorate has also been detected in tobacco (Nicotiana tabacum L.) (Ellington et al., 2001) and food crops [cucumber, Cucumis sativus L., and soybean, Glycine max (L.) Merr.] (Yu et al., 2004), and up to 300 mg kg–1 fresh wt. (on the basis of fresh wet weight) perchlorate uptake was found in salt cedar (Tamarix ramosissima Ledeb.) in Las Vegas Wash (Urbansky et al., 2000).
All of the above research demonstrates that perchlorate is accumulated mainly in leaves rather than in roots. However, to date, the majority of research has been conducted in well-controlled laboratory conditions for relatively short durations. Less information is available on seasonal uptake and release of perchlorate at field scale. Laboratory studies have implied that phytodegradation in plant tissues occurred fairly slowly, following a pathway of stepwise reduction 
(Nzengung et al., 1999; Aken and Schnoor, 2002). Slow transformation in plant tissues may raise concerns over the final fate of perchlorate in deciduous tree leaves. In addition, perchlorate uptake in the field may be affected by numerous factors, such as accessibility of perchlorate for uptake (shallow aquifer water table, distance to streams, and soil contamination level), exposure duration, species, dissolved solids content, and others.
The purpose of this study was to investigate perchlorate uptake in aquatic and terrestrial macrophytes from multiple streams surrounding a perchlorate-contaminated site, the Naval Weapons Industrial Reserve Plant (NWIRP), at McGregor, TX. This investigation included: (i) one year of aquatic plant and bulk water sampling (e.g., smartweed and watercress) from multiple streams, (ii) short-term sampling of terrestrial plants (e.g., willow and elm) from multiple streams, (iii) one year of seasonal leaf sampling (from budding to leaf drop) of a terrestrial plant habitat along a defined stream reach with a consistent perchlorate concentration, and (iv) redistribution or release potential of perchlorate in deciduous leaves trapped in leaf litter traps from this reach.
|
MATERIALS AND METHODS
|
|---|
Study Site
The NWIRP was associated historically with the use of solid propellant containing perchlorate. This site (approximately 4000 ha) is located 180 km south of Dallas, TX, between the Lake Waco and Lake Belton watersheds. Major local on-site and off-site watersheds at this site consist of Harris Creek, Willow Creek, South Bosque River, Onion Creek, and Station Creek. Harris Creek, Willow Creek, and the South Bosque River eventually discharge into Lake Waco. Station Creek and Onion Creek discharge to Lake Belton. Perchlorate contamination has been identified in soil, ground water, and multiple streams and creeks flowing away from the site (Hare, 2000). Soils at this site were reported to contain 23 to 1800000 µg kg–1 perchlorate. Both in situ (i.e., biobarriers with organic amendments) and ex situ technologies (fixed-bed anaerobic bioreactors) have been evaluated or adopted at this site to clean up the contaminated soil and ground water (Logan, 2001).
Water, Vegetation, and Soil Sampling
Six locations surrounding NWIRP were sampled, HW84 Mainstream, HW84 Sidestream, HW317, HW317/MN, HW107, and Mother Neff Road (Fig. 1). These sampling locations were distributed along three major creeks flowing away from the site, including Harris Creek, South Bosque River, and Station Creek. HW84 Mainstream and HW317 were located along Harris Creek, with annual water depth fluctuation of approximately 0.3 to 1.5 m. HW84 Sidestream was a 250-m-length spring-fed tributary of Harris Creek, with a relatively constant annual flow rate of about 4 x 10–3 m3 s–1. HW317/MN was located at an unidentified tributary to the South Bosque River, with a water depth fluctuation of about 0 to 0.6 m yearly. HW107 and Mother Neff Road were located at Onion Creek and South Bosque River, respectively.
Water and vegetation sampling were conducted from June 2001 through November 2002, at an interval of every 1 to 2 mo. Initially, short-term sampling was conducted for all the sites. Then, over a 12-mo period, sampling was conducted to focus on the remaining sites except HW107 and Mother Neff Road. Two dominant aquatic plants, smartweed and watercress, were collected. Two willow trees located at HW84 Sidestream and HW317/MN were also sampled.
In addition, starting from April 2002, seasonal tree leaf sampling was conducted for more than 40 trees, including ash, china-berry, elm, mulberry, and hackberry, which grew on a forestland (approximately 75 m long and 25 m wide) of a defined reach of HW84 Sidestream. Trees along the reach were tagged, and their locations were recorded. Tree leaves were sampled every 2 mo during the growing period of macrophytes (from budding to leaf drop). Tree leaves at different heights and branches of a tree were sampled with a leaf sampler and then placed together into a prelabeled plastic bag. At the end of October 2002, four leaf litter traps (1.2-m length x 1.2-m width) were deployed at four locations to collect deciduous leaves. Deciduous leaves were collected after all leaves had fallen from the trees. Distribution of trees sampled and the locations of litter traps are presented in Fig. 2.
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 2. Distribution of trees sampled along a defined reach of HW84 Sidestream (not to scale). All tress were tagged with different numbers.
|
|
Bulk water samples were taken using 30-mL plastic sample vials. Aquatic plants were rinsed with deionized water and then stored in plastic zip bags. All samples were put on ice, transferred to the laboratory, and stored in the refrigerator before analysis.
Soil samples (at depths of 15, 30, 60, and 75 cm below the surface) were taken in July 2003 at multiple locations approximately 1.5 m away from the stream bank of HW84 Sidestream with a hand auger (Fig. 1). Soil samples were put in plastic bags and transferred to the laboratory on ice for further extraction and analysis.
Analytical Methods
Water samples were filtered with IC Acrodisc 25-mm syringe filters (0.45-µm pore size; Pall Corporation, Ann Arbor, MI), and analyzed for perchlorate directly using USEPA Method 314.0 (USEPA, 1999). Aquatic and terrestrial plants were extracted following the general method of Ellington and Evans (2000). Aquatic plants were rinsed with deionized water and blotted dry with paper towels. Plant tissues were cut into 0.5- to 1-cm pieces and dried in the refrigerator for several days. Approximately 0.5 g of dried plant tissues were weighed and placed in 30-mL centrifuge tubes, and 30 mL of deionized water were added. Centrifuge tubes were then tightly capped and placed in a boiling water bath for 1 h, cooled to room temperature, and then placed in a 3°C refrigerator for 1 d. Samples were shaken regularly. Five milliliters of supernatant out of the total 30 mL were collected with a pipette and added to 5 g of precleaned aluminum oxide adsorbent (Al2O3, surface area = 155 m2 g–1, 150 mesh; Aldrich Chemical Company, Milwaukee, WI) for 1 d at 3°C. Samples were filtered with prefilters (Millipore, Billerica, MA) and IC Acrodisc 25-mm syringe filters (0.20-µm pore size; Pall Corporation), diluted with 18 M
water (1:5), and analyzed for ClO–4.
Soil samples were dried in a 105°C oven for 2 h. Approximately 5 g (dry weight) of homogenized oven-dried soil was placed in individual 50-mL flasks, and then 25 mL of deionized water were added to each flask. The flasks containing soil and water were shaken for 2 h at room temperature. The turbid soil–water mixture was then centrifuged. The supernatant was decanted and filtered using 0.20-µm IC Acrodisc 25-mm syringe filters before ClO–4 determination.
A Dionex (Sunnyvale, CA) DX-500 ion chromatograph was used to determine ClO–4 concentrations in accordance with USEPA Method 314.0 (USEPA, 1999). Perchlorate analysis was performed using IonPac AG16/AS16 guard and analytical columns. An anion self-regenerating suppressor (ASRS) with autosuppression at 100 mA was used for suppressed conductivity detection.
|
RESULTS AND DISCUSSION
|
|---|
Perchlorate Concentration in Surface Water
Perchlorate concentration in surface water was site-specific and temporally variable (Fig. 3). The highest perchlorate concentrations (up to 536 µg L–1) were detected at HW317/MN. Compared with the other three sites, the HW317/MN location had the highest ClO–4 concentration in the stream, probably because this location is close to an explosives disposal area within NWIRP (Hare, 2000). Perchlorate concentrations at HW84 Sidestream were relatively stable from January 2002 to October 2002 (20–40 µg L–1), but a peak of 123 µg L–1 concentration was observed in October 2001. The initial source of HW84 Sidestream is a spring. Both HW84 Sidestream and HW317/MN consistently contained ClO–4. Considerable fluctuation in perchlorate concentrations was observed at HW84 Mainstream, but perchlorate was detected in most months (11 of 16 samples). Perchlorate concentrations were below the detection limit (4 µg L–1) in August and September 2001 and August 2002 and were linked with rapid bacterial activity in natural wetland habitats at higher temperature. Significant variation in ClO–4 concentrations was also observed at HW317. It was evident that both HW84 Mainstream and HW317 intermittently received influxes of ClO–4. The variation of ClO–4 concentration may also have been caused by seasonal flow rate fluctuation in the streams due to different temporal evaporation and precipitation rates. On 6 Apr. 2002, surface water samples were collected at HW84 Mainstream and HW317 immediately after a heavy thunderstorm, so that very low perchlorate concentrations were detected. HW317 was located approximately 3 km downstream from HW84 Mainstream and exhibited a similar temporal trend in ClO–4 concentrations.
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 3. Perchlorate concentrations in surface water at different stream sites from June 2001 through October 2002. Points not connected by solid lines indicate that data are missing because water samples were not taken.
|
|
Perchlorate Accumulation in Aquatic Plants
Overall, ClO–4 accumulation in smartweed was observed at all the selected sites where ClO–4 was detected in surface water, except HW84 Mainstream (Table 1). Although the ClO–4 concentration in water averaged 12 µg L–1 at HW84 Mainstream (Table 1), ClO–4 may have been depleted in sediment pore water due to microbial degradation in the sediments when smartweed samples were collected, so that ClO–4 accumulation was not observed. A previous kinetics study demonstrated the rapid microbial degradation in sediment of HW84 Mainstream (intrinsic ClO–4 degradation rate constant k = 0.14 d–1) (Tan et al., 2004). Sediment pore water concentrations monitored by in situ dialysis samplers from another study indicated that ClO–4 was not present in sediment at HW84 Mainstream (Tan, 2003), and the absence of ClO–4 in sediment pore water supported this hypothesis. In addition, higher accumulation in smartweed was generally observed at sites with higher average ClO–4 surface water concentrations. Average ClO–4 accumulation in smartweed ranged from 9140 to 40600 µg kg–1 dry wt. for sites HW317/MN, HW84 Sidestream, and HW107, much higher than the range observed at HW84 Mainstream, HW317, and Mother Neff Road (<detection limit to 3180 µg kg–1) (Table 1). Watercress was collected from sites HW84 Sidestream, HW84 Mainstream, and HW107. Average perchlorate accumulation in watercress ranged from 625 to 4860 µg kg–1 dry wt. (Table 1).
View this table:
[in this window]
[in a new window]
|
Table 1. Perchlorate concentrations in streams and aquatic and terrestrial plants at multiple sites surrounding the Naval Weapons Industrial Reserve Plant (NWIRP), McGregor, TX. Samples were collected from June 2001 through October 2002.
|
|
Perchlorate accumulation in aquatic plants was regressed against ClO–4 bulk water concentration in streams. There was a significant linear relationship between ClO–4 concentration in bulk water and plant tissue of smartweed (n = 19, r2 = 0.7410, P 
0.0001) and watercress (n = 30, r2 = 0.6022, P 0.0001) (Fig. 4 and 5). There were no data points in the range of bulk water concentrations of 100 to 200 µg L–1 (Fig. 4) and 30 to 60 µg L–1 (Fig. 5), due to the availability of aquatic plants at the study sites during the sampling period. Thus, the slopes of lines in Fig. 4 and 5 were weighted by just a few points to the right-hand side of the axis, which resulted in the high statistical significance of the line. Results indicated that plant accumulation of ClO–4 from streams was substantial with dry leaf concentrations up to two orders of magnitude greater than bulk water concentrations. Bioconcentration factor (BCF), the ratio of contaminant concentration (mg kg–1) in the organism (plants, microorganisms, animals) to that in the water (mg L–1), was used to characterize the accumulation of perchlorate in plants. From the slope, the plant BCF on the basis of plant dry weight was estimated to be 164 for smartweed and 267 for watercress, respectively. A laboratory study by Susarla et al. (2000) indicated that as high as 456 mg kg–1 fresh wt. (based on plant fresh weight) ClO–4 uptake was taken up by smartweed exposed to 20 mg L–1 ClO–4 during a 10-d exposure experiment. Assuming 70% water content in smartweed, the BCF of Susarla's study (2000) was estimated to be about 76 (on the basis of plant dry weight), which is slightly lower than our results. Ten days may not be sufficient to achieve equilibrium of ClO–4 accumulation and the concentration tested is unrealistic compared with most surface water ClO–4 concentrations.
View larger version (34K):
[in this window]
[in a new window]
|
Fig. 4. Relationship between ClO–4 bulk water concentration and plant tissue concentration at different sampling locations for smartweed (n = 19, P 0.0001).
|
|
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 5. Relationship between ClO–4 bulk water concentration and plant tissue concentration at different sampling locations for watercress (n = 30, P 0.0001).
|
|
Compared with watercress, smartweed has a more dense and longer root system that can extend deep into the sediment, so smartweed may take up perchlorate primarily from sediment pore water. Watercress is a floating aquatic plant with short roots that can take up perchlorate directly from surface water and shallower sediment pore water. Thus, the BCF estimated on the basis of the bulk surface water perchlorate concentration instead of the pore water concentration may underestimate the accumulation potential of smartweed, because the pore water concentration is typically much lower than the bulk water concentration due to bacterial degradation in sediments. Because the actual sediment pore water concentrations at the locations where smartweed samples were collected were not available, bulk water concentrations were used to estimate the BCF in smartweed. Therefore, smartweed might have a higher perchlorate accumulation potential than watercress even if the BCF of smartweed estimated on the basis of the bulk water concentration is lower than that of watercress. As a result, observed average perchlorate accumulation in smartweed at HW84 Sidestream and HW317 was relatively higher than that of watercress (Table 1). At HW84 Mainstream, there was no perchlorate accumulation in smartweed, due to the absence of perchlorate in the sediment pore water (Fig. 4). In contrast, there was 4860 µg kg–1 in watercress because perchlorate was present in the bulk water for accumulation, stressing the importance of spatial exposure (Table 1 and Fig. 5).
Perchlorate Accumulation Potential in Different Species of Terrestrial Plants
Terrestrial plant leaf samples were collected from two sites with consistent perchlorate concentrations, HW84 Sidestream and HW317/MN. Perchlorate accumulation in all trees within 5 m from the stream was averaged to evaluate overall perchlorate accumulation potential of different species (Table 1). All tree species investigated (ash, china-berry, elm, mulberry, hackberry, and willow) showed significant potential for perchlorate accumulation, although perchlorate accumulation was variable and species dependent. Highest perchlorate accumulation was observed in hackberry (48200 µg kg–1 dry wt.), willow (24300 µg kg–1 dry wt.), and elm (14200 µg kg–1 dry wt.) (Table 1), implying that these species may be good candidates for phytoremediation. In addition, perchlorate was preferentially accumulated in the leaf of china-berry and mulberry trees (average 5040 and 1310 µg kg–1 dry wt., respectively) rather than the fruit (0 and 467 µg kg–1 dry wt., respectively) (Table 1), implying that perchlorate was selectively partitioned in these plants.
Seasonal Accumulation of Terrestrial Plants
Sampling of tree leaves during the growing season at the HW84 Sidestream site indicated that perchlorate accumulation in terrestrial plants was temporally and spatially variable and dependent on exposure duration. Generally, terrestrial plants up to 9 m from the stream accumulated significant concentrations of perchlorate (Fig. 6). Trees located closer to the stream seemed to have higher perchlorate accumulation, but in some cases trees located farther away from the stream also exhibited high perchlorate accumulation. This variation may be caused by the heterogeneous distribution and penetration of root systems of trees. Some trees located further away from the stream may have widely extended root systems to absorb water from the stream. For most trees, low leaf ClO–4 concentrations were detected in April and June 2002, and the highest leaf tissue concentrations were observed late (August and October 2002) in the growing cycle (Fig. 6). Seasonal sampling of a willow tree from another site (HW317/MN) also indicated that perchlorate accumulation in leaf tissue was seasonally variable (Table 2). Perchlorate accumulation in the willow tree increased with progression through the growth cycle. In April and May 2002, there was no perchlorate accumulation observed in new buds. In June and August, 2002, perchlorate accumulation in leaves was 300 ± 210 and 1200 ± 250 µg kg–1 dry wt., respectively. The highest accumulation in leaves was detected late in the growing cycle (15800 ± 8200 and 14600 ± 11400 µg kg–1 dry wt. for October 2001 and October 2002, respectively).
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 6. Perchlorate seasonal distribution in tree leaves along a defined reach of HW84 Sidestream. Standard deviation bars represent the standard deviation of duplicates (n = 2) for selected trees sampled in October 2002. In cases where error bars are not shown, there is no duplicate.
|
|
The reason that perchlorate concentrations in some trees showed declining trends in October 2002 compared with that in August 2002 is not clear (Fig. 6). One possibility was that perchlorate may retranslocate and move back from annual leaves to perennial woody stem tissues, jut like other macronutrients and micronutrients. Sampling woody tissues might be able to yield useful information. Another possibility was that the leaching and rainfall may decrease the perchlorate concentration in leaves, especially for those tree leaves starting to fall off during the late growing season. Phytodegradation may contribute to a partial decrease of perchlorate in plant tissues, but this process normally occurs relatively slowly (Nzengung et al., 1999; Aken and Schnoor, 2002). Nzengung et al. (1999) estimated that phytodegradation accounted for approximately 11% ClO–4 loss in a 26-d laboratory growing experiment using willow trees.
The mechanisms of perchlorate transport in plants are not fully understood. Perchlorate is a nonvolatile and highly soluble anion, which is not readily adsorbed by the negatively charged surface of most soils (Logan, 2001). Perchlorate transport in plants might be linked with passive transport across the plasma membranes of root cells (i.e., simple diffusion or facilitated diffusion). Under this assumption, perchlorate uptake may be dependent on mass flow or transpiration. There are plant uptake data for contaminants that should behave similarly to perchlorate in the environment, due to similar atomic radius 
(Cataldo et al., 1983; Krijger et al., 2000). These data indicate plant uptake and possible transporter, as well as specific relationships (competition) to other ionic species like nitrate, sulfate, and phosphate, but it is not known whether passive or active transport is involved. Further studies are needed to understand the mechanisms of perchlorate transport in plants.
Another hypothesis is that the availability of other ClO–4 sources, such as contaminated soil and ground water, may also affect the accumulation and distribution of ClO–4 in terrestrial plants. However, information obtained from the HW84 Sidestream site suggests that terrestrial plants grown on this site absorb ClO–4 mainly from the stream. In general, ClO–4 was not detected in soil samples up to a depth of 60 cm beneath the surface. Only 1 out of 11 soil samples had a ClO–4 concentration of 0.03 mg kg–1 dry wt. Results obtained suggested that soils near HW84 Sidestream were not contaminated by ClO–4. The boring log of a ground water monitoring well located near HW84 Sidestream was composed of brown silty clay with limestone cobbles (0–1.2 m), slightly soft and moist clay (1.2–2.3 m), hard and silty argillaceous limestone and shale (2.3–3.0 m), and dry dark gray shale (3.0–5.2 m). Perchlorate concentrations were below the detection limit (4 µg L–1) in ground water sampled from the monitoring well from May 2000 to July 2001 (MWH Global, personal communication, 2003). This information suggested that trees at this site take up ClO–4 mainly from the stream instead of from ground water sources. An attempt was made to correlate tree size with ClO–4 accumulation, but no correlation was found. Tree size may affect ClO–4 accumulation in case of vertical contaminant source present (i.e., contaminated ground water or soil), because bigger and taller trees tend to have a deeper root extension. However, in this case, ClO–4 accumulation in terrestrial plants was mainly influenced by the distance to the stream, since only the horizontal ClO–4 source (stream) was available for accumulation. On the other hand, trees with large trunks and wide root systems could also have a relatively high ClO–4 accumulation even if they were located farther away from the contaminated stream.
Perchlorate concentrations in deciduous tree leaves collected from four litter traps (layout shown in Fig. 2) in January 2003 were compared with concentrations in live tree leaves sampled before leaf drop in October 2002. Average concentration of trees located in the proximity of litter traps (within a radius of 1.5 m away from individual litter traps) was considered to be a typical ClO–4 concentration in the late growth cycle. Generally, average ClO–4 leaf tissue concentrations before leaf drop were higher than that in tree leaves after leaf drop for individual litter traps (Table 3). It was statistically significant (t test, p 0.05) that ClO–4 concentrations in deciduous tree leaves in Litter Traps 3 and 4 were lower than those in live leaves before leaf drop (Table 3). However, in some cases (e.g., in Litter Traps 1 and 2), according to the statistical analysis (t test), there was no significant difference in ClO–4 concentration before and after leaf drop (t test, p > 0.05).
The decrease of perchlorate concentration in deciduous leaves was most likely associated with leaching, rainfall, and microbial degradation. Perchlorate from plants may be released into the environment, suggesting that plants may serve as source (re-release) and sink (accumulation) of perchlorate. The selection of suitable plants may become a key factor in successful phytoremediation. The potential of evergreen plants to remediate perchlorate needs to be considered. If deciduous plants are to be used in phytoremediation of perchlorate, harvest and subsequent disposal after the plants hyperaccumulate perchlorate may be necessary.
|
CONCLUSIONS
|
|---|
This research elucidated the fate of ClO–4 in macrophytes in natural systems using the NWIRP site as a case study. Perchlorate concentrations in surface waters at multiple streams were temporally variable, depending on numerous factors such as contamination source, fluctuation of flow rate, and bacterial degradation. Significant ClO–4 uptake was observed in smartweed and watercress that dominated the natural wetland habitat. Perchlorate uptake in leaves of terrestrial plants was dependent on numerous factors, such as plant species, accessibility to the ClO–4 source, contamination levels in soils and ground water, as well as exposure duration. Perchlorate taken up by trees may re-release to the environment after uptake. Results indicated that terrestrial plants at this site mainly take up perchlorate from streams rather than from ground water. However, at some sites with a shallow ground water table, if terrestrial plants absorb and accumulate ClO–4 from ground water and the BCF of a specific plant species is known, terrestrial plants may serve as an another alternative to monitor ClO–4 contamination in ground water by simply determining the ClO–4 uptake in terrestrial plants overlying the ground water table. Tree size may become an important factor to affect the uptake of ClO–4, since bigger and taller trees tend to have a deeper and more dense root system capable of reaching the ground water. Thus, terrestrial plants could become a useful biomonitoring tool as a supplement to costly ground water monitoring wells. Information obtained will be helpful to tailor a site-specific design and management protocol of phytoremediation to remediate ClO–4 contamination in other similar sites with nonpoint ClO–4 contamination sources.
|
ACKNOWLEDGMENTS
|
|---|
This research was funded by the Brazos River Authority through the U.S. Army Corps of Engineers (Fort Worth District).
|
REFERENCES
|
|---|
- Aken, B.V., and J.L. Schnoor. 2002. Evidence of perchlorate ClO–4 reduction in plant tissues (poplar tree) using radio-labeled 36ClO–4. Environ. Sci. Technol. 36:2783–2788.[Medline]
- Cataldo, D.A., R.E. Wildung, and T.R. Garland. 1983. Root absorption and transport behavior of technetium in soybean. Plant Physiol. 73:849–852.[Abstract/Free Full Text]
- Damian, P., and F.W. Pontius. 1999. From rockets to remediation: The perchlorate problem. Environmental Protection. June, p. 24–31.
- Ellington, J.J., and J.J. Evans. 2000. Determination of perchlorate at parts-per-billion levels in plants by ion chromatography. J. Chromatogr. A 898:193–199.[ISI][Medline]
- Ellington, J.J., N.L. Wolfe, A.W. Garrison, J.J. Evans, J.K. Avants, and Q. Teng. 2001. Determination of perchlorate in tobacco plants and tobacco products. Environ. Sci. Technol. 35:3213–3218.[Medline]
- Hare, J.P. 2000. Remediation of perchlorate contamination at NWIRP McGregor, Texas. Federal Facilities Environmental Journal. Winter, p. 33–37.
- Hunter, W. 2002. Bioremediation of chlorate or perchlorate contaminated water using permeable barriers containing vegetable oil. Curr. Microbiol. 45:287–292.[ISI][Medline]
- Krijger, G.C., C. Kolloffel, and H.T. Wolterbeek. 2000. Effect of nitrate on uptake of pertechnetate by tomato plants. J. Environ. Qual. 29:866–870.[Abstract/Free Full Text]
- Logan, B.E. 2001. Assessing the outlook for perchlorate remediation. Environ. Sci. Technol. 12:483A–487A.
- Motzer, W.E. 2001. Perchlorate: Problems, detection, and solutions. Environ. Forensics 2(4):301–311.
- Na, C.Z., F.S. Cannon, and B. Hagerup. 2002. Perchlorate removal via iron-preloaded GAC and borohydride regeneration. J. Am. Water Works Assoc. 94(11):90–102.
- Nerenberg, R., B.E. Rittmann, and I. Najm. 2002. Perchlorate reduction in a hydrogen-based membrane-biofilm reactor. J. Am. Water Works Assoc. 94(11):103–114.
- Nzengung, V.A., C. Wang, and G. Harvey. 1999. Plant-mediated transformation of perchlorate into chloride. Environ. Sci. Technol. 33:1470–1478.
- Susarla, S., S.T. Bacchus, G. Harvey, and S.C. McCutcheon. 2000. Phytotransformations of perchlorate contaminated waters. Environ. Technol. 21:1055–1065.
- Susarla, S., S.T. Bacchus, N.L. Wolfe, and S.C. McCutcheon. 1999. Phytotransformation of perchlorate and identification of metabolic products in Myriophyllum aquaticum. Int. J. Phytorem. 1:97–107.
- Tan, K. 2003. Fate of perchlorate in natural systems: Intrinsic biodegradation, plant uptake, and remediation potential of wetlands. Ph.D. diss. Dep. of Civil Eng., Texas Tech Univ., Lubbock.
- Tan, K., T.A. Anderson, and W.A. Jackson. 2004. Degradation kinetics of perchlorate in sediments and soils. Water Air Soil Pollut. 151:245–259.
- Urbansky, E.T. 1998. Perchlorate chemistry: Implications for analysis and remediation. Biorem. J. 2(2):81–95.
- Urbansky, E.T., M.L. Magnuson, C.A. Kelty, and S.K. Brown. 2000. Perchlorate uptake by salt cedar (Tamarix ramosissima) in the Las Vegas Wash riparian ecosystem. Sci. Total Environ. 256:227–232.[Medline]
- USEPA. 1999. Method 314.0. Determination of perchlorate in drinking water using ion chromatography. Revision 10. Doc. 815-B-99-003. USEPA, Washington, DC.
- Yu, L., J.E. Cãnas, G.P. Cobb, W.A. Jackson, and T.A. Anderson. 2004. Uptake of perchlorate in terrestrial plants. Ecotoxicol. Environ. Saf. 58:44–49.[ISI][Medline]
Related articles in JEQ:
- This Issue in Journal of Environmental Quality
JEQ 2004 33: 1589-1599.
[Full Text]
This article has been cited by other articles:
|
|
|
|
 
D. D. Bannerman, M. J. Paape, R. L. Baldwin VI, C. P. Rice, K. Bialek, and A. V. Capuco
Effect of mastitis on milk perchlorate concentrations in dairy cows.
J Dairy Sci,
August 1, 2006;
89(8):
3011 - 3019.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
