| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Texas Agricultural Experiment Station, Texas A&M University, 17360 Coit Road, Dallas, TX 75252-6599
b Dep. of Soil, Water and Climate, Univ. of Minnesota, 1991 Upper Buford Circle, St. Paul, MN 55108
c Metropolitan Council Environmental Services, 2400 Childs Road, St. Paul, MN 55106
* Corresponding author (j-sloan{at}tamu.edu)
Received for publication January 8, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARYREFERENCES |
|---|
Abbreviations: CVAFS, cold vapor atomic fluorescence spectroscopy
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARYREFERENCES |
|---|
Additions of wastewater biosolids (also known as sewage sludge) to agricultural soils can significantly increase the amount of Hg relative to background concentrations—especially when the soil receives multiple applications (Steinnes, 1990). Mean background concentrations of Hg in soils near St. Paul, Minnesota, were measured at 13 and 48 µg kg-1 for a sandy agricultural soil and a forested soil, respectively (RUST Environment & Infrastructure, 1994). A national sewage sludge survey conducted in 1988 determined that the median concentration of Hg in wastewater biosolids was 1.5 mg kg-1 dry sludge (USEPA, 1990). Mercury concentrations in biosolids from the Metropolitan Water Reclamation District of Greater Chicago ranged from 1.1 to 8.5 mg kg-1 during the years 1971 to 1994 (Granato et al., 1995). A one-time application of a much cleaner biosolids material (0.1 mg Hg kg-1) at an agronomic rate of 10 Mg ha-1 would result in a Hg application of 100 µg m-2. This conservative estimate is 10 to 20 times greater than annual atmospheric deposition of Hg. Consecutive applications of biosolids to soil can greatly increase the Hg content of the soil (Williams et al., 1985). Soil organic matter plays an important role in retaining Hg in soil (Barrow and Cox, 1992; Lodenius et al., 1987; McNab et al., 1997; Yin et al., 1996). Most of this Hg remains near the soil surface (Granato et al., 1995; McNab et al., 1997), where it is more susceptible to erosional processes.
Two factors are primarily responsible for the lack of information on Hg in agricultural watersheds. First, Hg has generally been excluded from the list of metals studied in biosolids amended–soils research. The exclusion of Hg is due in part to the need for separate analytical methods and instrumentation for its determination. Second, the focus of most research on land application of biosolids has been the transfer of trace metals to the food chain via plant uptake (Dowdy et al., 1991; Dowdy et al., 1983a,b; Giordano et al., 1975; Hinesley et al., 1976; Page et al., 1987). Plants take up very little Hg from soils and potential contamination of the food chain through that pathway is low (Chaney, 1990). Most Hg found in aboveground plant parts results from adsorption of volatile Hg forms (Lindberg et al., 1979). There is little literature focusing on the movement of Hg in agricultural watersheds (Balogh et al., 2000), especially where biosolids were applied.
Since the USEPA (1993) promotes beneficial use of biosolids in agriculture, and since biosolids can greatly increase the Hg content of soils, it is essential that we know the fate and transport of Hg in agricultural systems. The objective of this study was to evaluate the effect of 20 consecutive years of biosolids applications to agricultural soils on Hg accumulation in soil and its transport into surface runoff water.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARYREFERENCES |
|---|
|
Each year that biosolids were applied, a representative sample was collected and freeze-dried for storage. Unfortunately, Hg was not determined in the original biosolids samples and stored samples were only found for the years 1974, 1978, and 1990. These samples were analyzed for total Hg to estimate cumulative Hg loadings.
Watershed terraces were designed so that excess water would flow toward natural low points in the terrace channel where vertical surface inlet pipes directed flow through underground solid-wall PVC pipe to a sampling station at the lower end of the watershed. After the sampling station, water is discharged into a grassy waterway where it infiltrates into the soil and/or flows over ground to a runoff collection lagoon, depending on the intensity of the runoff event. Two, and in the case of snowmelt runoff, three terraces were used to investigate the effect of cumulative biosolids applications to cultivated watershed soils on Hg transport and deposition (Fig. 1). The three treatments areas were (i) no biosolids (west end of Terrace 4), (ii) 87 Mg ha-1 biosolids (west end of Terrace 3), and (iii) 224 Mg ha-1 biosolids (Terrace 5).
Protocols for Mercury Sampling and Analysis
Sampling and laboratory techniques are very important when working with Hg due to the potential for contamination. Strict quality assurance/quality control (QA/QC) procedures were used during all phases of this project to assure collection of contaminant-free samples that would provide accurate and reliable Hg data. Collection of soil and water samples was conducted using a "clean-hands/dirty-hands" technique (Ahlers et al., 1990). With this technique, only the "clean-hands" person, wearing clean-room garb, touches the actual sample and sample container. Mercury analyses were conducted in a clean-room laboratory that had a USEPA-accepted Quality Assurance Project Plan.
Sampling and Analysis of Watershed Soils and Plants
Soil samples were collected from the control and the 224 Mg ha-1 biosolids-treated terraces along two transects. Both transects originated at the surface inlet, but one ran on the contour, parallel to tillage operations, and the other ran perpendicular to tillage (Fig. 1). The rationale of the soil sampling scheme was to determine if there had been translocation of biosolids-applied Hg toward the surface inlet, which was the focus of runoff in each terrace. Soil samples were collected from the 0- to 15- and 15- to 30-cm depths at four points along each transect. The distance between sampling points was increased as the distance from the surface inlet increased. Three subsamples were collected from a 1-m2 area at each sampling point. The three subsamples were homogenized in an acid-washed polyethylene bucket using a teflon-coated spatula and then transferred to a double-lined plastic bag. Chemical and physical properties of the soil were determined using established procedures. Organic carbon was determined by measuring the amount of CO2 released during dry combustion (Nelson and Sommers, 1982). Particle size distributions were determined by the pipette method (Gee and Bauder, 1986). Soil pH was measured at a 1:1 soil to water ratio (McLean, 1982). Total Hg was determined on all soil samples using procedures described below.
Since 1973, when the watershed terraces were constructed, a considerable amount of runoff has been discharged onto the grassy, noncultivated waterway near the center of the watershed, and from there, into a lagoon (Fig. 1). Mercury transported in this runoff may potentially affect aquatic environments, such as rivers, lakes, or wetlands. If Hg levels were elevated in sediments deposited in and around the runoff lagoon, then vegetation near the lagoon would possibly contain elevated levels of Hg due to vascular uptake of soil Hg and/or stomatal uptake of volatilized Hg0. To test this possibility, reed canarygrass samples were collected in July 1996 from the edge of the lagoon where runoff from the terraces was collected. Grass samples were also collected from another area approximately 100 m from the edge of the lagoon that was not affected by runoff waters. At each sample site, three aboveground grass samples (leaves plus stems) were collected from an area approximately 10 m2. Each sample was treated as a replication. Plant tissue was digested in a similar fashion as the soil samples and analyzed for total Hg.
Runoff Sampling, Processing, and Analysis
During the spring of 1996, there was a single snowmelt event at the Rosemount watershed that lasted four consecutive days. Snowmelt runoff samples were manually collected each day using the "clean hands/dirty hands" technique (Ahlers et al., 1990). Water samples were collected in 125-mL Teflon bottles that were washed in hot 50% HNO3, rinsed in ultrapure water, filled with 1% ultrapure HNO3, and sealed inside two polyethylene bags. Sample bottles were stored and transported to the sampling site in closed acid-cleaned plastic tubs (clean-boxes). Snowmelt samples were collected by hand and transported to the laboratory for analysis. Snowmelt runoff rates were measured at the same time the water samples were collected by recording the time required to fill a 19-L polyethylene container. In the laboratory, runoff samples were split into two parts and half the sample was filtered through a 0.45-µm cellulose nitrate membrane filter. Total Hg was measured in both the unfiltered and filtered samples to distinguish between particulate-bound and dissolved Hg.
Mercury Analysis
Analytical determination of Hg is difficult due to low concentrations and the high potential for contamination. However, recent improvements in analytical techniques have made it possible to determine sub-picogram quantities of Hg in samples (Bloom, 1989). Several techniques were used to assure analytical quality control. Multiple spiked samples were analyzed to determine percent recovery. Average percent recovery was 94 ± 5%. Standard reference material (SRM 2704, Buffalo River Sediment) was analyzed to assess accuracy of the methods. The average value for multiple analyses of SRM 2704 was within the 95% confidence interval of the certified value. Reagent blanks were routinely analyzed to check for contamination.
Total Mercury in Soils, Plants, Sediments, and Biosolids
Solid materials were digested in a Teflon bomb with concentrated HNO3. An appropriate amount of sample solids (0.2 to 1 g) were placed in a Teflon bomb with 3 to 5 mL concentrated HNO3 and heated at 140°C for 3 h in a laboratory oven (Van Delft and Vos, 1988). The sample was allowed to cool before opening and the entire contents of the Teflon container were quantitatively transferred to a 25-mL volumetric flask, reduced by the addition of hydroxylamine hydrochloride, and diluted with water. Mercury in the digest was determined by cold vapor atomic fluorescence spectroscopy (CVAFS).
Total Mercury in Water
A cold digestion technique that utilizes the strong oxidizing capacity of bromine monochloride (BrCl) was used to completely oxidize organomercurial compounds so that total Hg could be measured in runoff water samples (Bloom and Crecelius, 1983). Water samples were routinely analyzed by adding an aliquot of BrCl (5 mL L-1), allowing it to react at room temperature for 24 h, and then measuring Hg in the digest by CVAFS.
Cold Vapor Atomic Fluorescence Spectrometry (CVAFS)
The only form of Hg that is detectable by CVAFS is the elemental form, Hg0 (Bloom, 1989). Soil and water digests contained inorganic forms of Hg, predominantly Hg2+. Inorganic Hg was reduced to Hg0 by adding 10% SnCl2 to an aliquot of the sample solution in a closed reaction flask. Volatile Hg0 was quantitatively carried from the reaction vessel in a stream of N2 gas and captured on a gold trap. The gold trap was transferred to the CVAFS system and Hg was determined using double gold trap amalgamation (Fitzgerald and Gill, 1979). The CVAFS system was able to accurately measure from <1 pg to >100 ng Hg.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARYREFERENCES |
|---|
|
|
In contrast to the control terrace, Hg concentrations in the surface soil (0–15 cm) of the biosolids-treated terrace were much higher than those in the subsurface (15–30 cm). The fact that subsurface soils in the biosolids-treated terrace had Hg concentrations that were 2 to 10 times higher than those for the control terrace indicate that there has been downward movement of Hg following biosolids applications. It is likely that downward movement of biosolids-applied Hg was in part due to tillage operations. Biosolids were initially injected to a depth of 10 to 15 cm at the time of application. Annual chisel plowing probably mixed Hg-enriched surface soil with unaffected subsoil. Deeper soil sampling is needed to determine if soil Hg concentrations at depths >30 cm are elevated above background concentrations. However, other researchers have reported that 80 to 100% of biosolids-applied Hg remained in the upper 10 to 30 cm of cultivated, agricultural soils (Granato et al., 1995; McNab et al., 1997). Williams et al. (1985) reported that soil Hg concentrations 5 cm below the zone of biosolids incorporation were equal to background levels.
The primary purpose for measuring soil Hg along transects radiating from surface inlets was to determine if biosolids-applied Hg had migrated toward the surface inlet due to erosional processes. Landscape slopes toward the inlets were generally steeper along transects that were perpendicular to tillage operations than those that were parallel. Mercury concentrations along control treatment transects (Fig. 3a,b) were similar (excluding the outlying point) and suggest a uniform distribution of Hg that might be attributed to natural sources as well as atmospheric deposition. Soil Hg concentrations generally increased with distance from the tile inlet along the parallel transect of the biosolids-treated terrace (Fig. 3d), but a linear trend was only significant for the 15- to 30-cm depth (r2 = 0.93, p < 0.05). There was no consistent trend in Hg concentrations along the perpendicular transect of the biosolids-treated terrace (Fig. 3c). Soil Hg concentrations along transects in both terraces were not correlated to soil texture or pH. There appeared to be a weak correlation between soil Hg and soil organic C (data not shown), but the relationship was only statistically significant for the 15- to 30-cm depth of the perpendicular transect on the control terrace (r2 = 0.987, p < 0.01). Research has demonstrated that soil organic matter plays an important role in retaining Hg in soil (Barrow and Cox, 1992; Lodenius et al., 1987; McNab et al., 1997; Yin et al., 1996). In our study, the strength of the statistical relationship between soil Hg and organic C might have been improved by increasing the number of sampling points per transect or by including additional transects.
Lower soil Hg concentrations near the surface inlet of the biosolids-treated terrace along the parallel sampling transect may be a result of complementing processes. First, as the tractor-pulled biosolids injector approached the surface inlet head-on (parallel direction), it could only pass within 5 to 10 m, but as it pulled to the side of the surface inlet (perpendicular direction), it was able to approach the inlet to within 1 to 2 m. During the 20 years biosolids were applied to this terrace, it is likely that cumulative loadings were lower near the surface inlet along the parallel transect. Soil erosion was another process that probably reduced soil Hg levels near the surface inlet. The topography of the biosolids-treated terrace was such that runoff occurred more often and quickly from areas adjacent to the surface inlet versus that from distances further away. The erosional loss of surface sediments around the inlet probably removed a disproportionately larger amount of biosolids-derived Hg. Soil Hg levels near surface inlets could be further reduced by in situ reduction of oxidized inorganic Hg to volatile Hg0 as a result of longer periods of saturated soil-water conditions in lower landscape positions where surface inlets drain ponded runoff water (Carpi et al., 1997).
Elevated concentrations of soil Hg at 20 and 30 m from the tile inlet along the parallel transect might also be due to erosional processes. From the time the terraces were constructed in 1973, there has been a gradual movement of sediments from the top of the terrace to the bottom of the terrace (i.e., perpendicular direction). The parallel terrace was located near the bottom of the terrace (Fig. 1) where Hg-bearing sediments might have accumulated.
Mercury in Snowmelt Runoff
Snowmelt runoff samples were collected during four consecutive days in March 1996 using clean sampling and analytical techniques. The rate of runoff varied among the three terraces sampled (Table 1), and depended on the surface topography of the terrace as well as vertical and horizontal distributions of snowdrifts. Runoff rates were highest for the south-facing 224 Mg ha-1 biosolids-treated terrace and lower for the north-facing control and 87 Mg ha-1 biosolids terraces.
|
Biosolids applications had no effect on Hg concentrations in filtered (<0.45 µm) snowmelt samples (Table 1). Mercury concentrations in the filtered samples ranged from 2 to 4.7 ng L-1 for the control and two biosolids-treated terraces. Low Hg concentrations in the filtered samples show that most Hg in snowmelt runoff was associated with particulates >0.45 µm in diameter. The lack of differences between the control and biosolids-treated areas suggests that dissolved Hg and Hg associated with particulates <0.45 µm in diameter were derived from naturally occurring or atmospherically deposited sources. Total Hg in the unfiltered and filtered samples were similar to those observed by Balogh et al. (2000) in snowmelt runoff from an agricultural watershed in an adjacent county. The total Hg concentrations are much lower than 2000 ng L-1, which is the current Maximum Contaminant Level for drinking water established by the USEPA (USEPA, 2001).
Mercury in Plant Tissue
Reed canarygrass growing near the edge of the watershed lagoon contained 13.4 µg kg-1 Hg, which was significantly higher than a concentration of 4.1 µg kg-1 in the same grass species growing in an area away from the lagoon that was not influenced by runoff from biosolids-amended terraces (Fig. 4)
. This comparison suggests that Hg from the biosolids-treated terraces was transported to a non-treated area of the watershed through hydrological processes and was taken up by vegetation, either through vascular tissue or as volatilized Hg0 through the stomata. The plant Hg concentrations in this study are similar to values reported in the literature. Granato et al. (1995) reported that 51% of corn leaf samples from soil that received 1317 Mg ha-1 biosolids had Hg concentrations <25 µg kg-1 (detection limit). Dowdy et al. (1983b) found that Hg levels in corn silage were below a detection limit of 50 µg kg-1 following a three-year cumulative Hg loading of 0.9 kg ha-1. On the other hand, Warman et al. (1995) found Hg concentrations of 100 to 300 µg kg-1 in Swiss chard (Beta vulgaris var. flavescens (Lam.) DC.] grown in a variety of composted biosolids.
|
|
SUMMARY |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARYREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARYREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. Tian, T. C. Granato, R. I. Pietz, C. R. Carlson, and Z. Abedin Effect of Long-Term Application of Biosolids for Land Reclamation on Surface Water Chemistry J. Environ. Qual., January 3, 2006; 35(1): 101 - 113. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Vadose Zone Journal | Journal of Plant Registrations | ||||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal |
