Environ. Sci. Technol. 2003, 37, 951-957
Biotransformation and Accumulation of Arsenic in Soil Amended with Seaweed H A Y L E Y C A S T L E H O U S E , †,‡ CASSANDRA SMITH,† ANDREA RAAB,† CLAIRE DEACON,‡ ANDREW A. MEHARG,‡ AND JO ¨ R G F E L D M A N N * ,† Department of Chemistry, University of Aberdeen, Meston Walk, Old Aberdeen, AB24 3UE, Scotland, U.K., and Department of Plant and Soil Science, University of Aberdeen, St. Marchar Drive, Old Aberdeen, AB24 3UE, Scotland, U.K.
For many coastal regions of the world, it has been common practice to apply seaweed to the land as a soil improver and fertilizer. Seaweed is rich in arsenosugars and has a tissue concentration of arsenic up to 100 µg g-1. These arsenic species are relatively nontoxic to humans; however, in the environment they may accumulate in the soil and decompose to more toxic arsenic species. The aim of this study was to determine the fate and biotransformation of these arsenosugars in soil using HPLC-ICPMS analysis. Data from coastal soils currently manured with seaweeds were used to investigate if arsenic was accumulating in these soils. Long-term application of seaweed increased arsenic concentrations in these soils up to 10-fold (0.35 mg of As kg-1 for nonagronomic peat, 4.3 mg of As kg-1 for seaweed-amended peat). The biotransformation of arsenic was studied in microcosm experiments in which a sandy (machair) soil, traditionally manured with seaweed, was amended with Laminaria digitata and Fucus vesiculosus. In both seaweed species, the arsenic occurs in the form of arsenosugars (85%). The application of 50 g of seaweed to 1 kg of soil leads to an increase of arsenic in the soils, and the dominating species found in the soil pore water were dimethylarsinic acid (DMA(V)) and the inorganic species arsenate (As(V)) and arsenite (As(III)) after the initial appearance of arsenosugars. A proposed decomposition pathway of arsenosugars is discussed in which the arsenosugars are transformed to DMA(V) and further to inorganic arsenic without appreciable amounts of methylarsonic acid (MA(V)). Commercially available seaweed-based fertilizers contain arsenic concentration between 10 and 50 mg kg-1. The arsenic species in these fertilizers depends on the manufacturing procedure. Some contain mainly arsenosugars while others contain mainly DMA(V) and inorganic arsenic. With the application rates suggested by the manufacturers, the application of these fertilizers is 2 orders of magnitude lower than the maximum permissible sewage sludge load for arsenic (varies from 0.025 kg ha-1 yr-1 in Styria, Austria, to 0.7 kg ha-1 yr-1 in the U.K.), * Corresponding author phone: +44-0-1224-272911; fax: +440-1224-272921; e-mail:
[email protected]. † Department of Chemistry. ‡ Department of Plant and Soil Science. 10.1021/es026110i CCC: $25.00 Published on Web 01/31/2003
2003 American Chemical Society
while a direct seaweed application would exceed the maximum arsenic load by at least a factor of 2.
Introduction Seaweed was once widely used as a fertilizer in coastal regions of the Atlantic seaboard of Europe and elsewhere access to this resource was available and arable farming was conducted (1-3). It is rich in a wide range of nutrients, improves soil texture and quality, and is a renewable and sustainable resource as seaweed is deposited routinely in large quantities on many beaches. For coastal farms, this resource is on their doorstep and requires only limited transportation. With the advent of modern fertilizers, it fell out of use because of the ease of application of formulated pesticides and because the contamination of seaweed washed up on beaches with plastic waste (fishing lines and ropes, bottles, etc.) led to a decline in use (2). With the onset of green agriculture and the increasing demand for organically farmed produce, seaweed use as a fertilizer has come back into fashion (3). In places such as the Outer Hebrides on the west coast of Scotland there are government subsidies for returning to traditional forms of agriculture to maintain floral and faunal diversity, and this involves reversion to the use of seaweed as manure. As well as seaweed being applied directly to soil, liquid preparations are sold for more intensive agriculture, such as greenhouse food production, again as the product is organic (4). One factor has been overlooked in this reversion to seaweed use. Seaweed naturally contains high levels of arsenic, typically between 20 and 100 mg kg-1 dry weight (dw) (5). Thus, sustained use of seaweed may lead to the buildup of arsenic in soils. The dominant species of arsenic in these seaweed are in the form of arsenoribofuranosides (arsenosugars). These are assumed to be relatively nontoxic to humans and animals as compared to inorganic species (6, 7). The arsenosugars are metabolized to different organoarsenic species but mainly to DMA(V) (dimethylarsinic acid) when consumed as a food source (8, 9). The species of arsenic has a direct effect on both its toxicity and mobility. It is not known how the arsenosugars in the seaweed are transformed after application to soil as a fertilizer. Therefore, it is not possible to determine the ecotoxicological effects and subsequent risk to human health resulting from such activities. The aim of this study was to determine the biotransformation of arsenic in soil amended with seaweed and to determine if sustained seaweed application led to substantial buildup of arsenic in soil.
Experimental Section Microcosm Experiment. (a) Seaweed. The two seaweed species that were chosen for use in the short-term experiment were Fucus vesiculosus and Laminaria digitata both Phaeophyta but from different algal families. These species were chosen because they are two of the most common species used as fertilizers in Scotland. Both species used were collected at Cowie near Stonehaven, Aberdeen, NE Scotland. The seaweed was collected a couple of days following a heavy storm. L. digitata used in this experiment was storm-cast weed deposited on the beach, and the F. vesiculosus was cut from the rocks at the same site on the beach as there were no storm-cast specimens available. (b) Seaweed Pretreatment. The seaweed was washed thoroughly with tap water. The specimens were then laid out on plastic crates and left to dry. Following the drying VOL. 37, NO. 5, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
951
period, the seaweed samples were hand-cut using scissors into uniform pieces. The L. digitata was cut into approximately 1-cm2 pieces, and the F. vesiculosus was cut into pieces approximately 2 cm in length. (c) Soil Samples. The soil used for analysis of the shortterm effects of seaweed fertilizer application was a fertile coastal vegetated sandy (locally called machair) soil collected from the west coast of Scotland. Machair soils are still widely cultivated in the Western Isles of Scotland, with seaweed being the traditional manure. All soils were 2 mm wet sieved prior to use and stored at 4 °C until they were analyzed. (d) Microcosm Design. The use of microcosms was chosen for the study of short-term effects of seaweed application on the soil. A total of 12 microcosms were set up. Each microcosm consisted of a 500-mL Erlenmeyer flask with Parafilm placed over the neck of the flask, this was to allow air to move in and out of the microcosm but to inhibit the movement of water, thus controlling the moisture content within the microcosm. In addition, the microcosms were kept in the dark to inhibit the growth of algae. All the microcosms contained 300 g of presieved soil, and each microcosm was subjected to one of three fertilizer treatments. In total, four replicates of each treatment was set up. F. vesiculosus and L. digitata were each added to one set of replicates at a rate of 50 g kg-1 (slightly higher than the field application rate of 25 t/ha). The other set of replicates was a control with no seaweed added. To all the microcosms, 50 mL of deionized water was added to bring the soil to 65% of the soils waterholding capacity. The soils in each of the microcosms were thoroughly mixed, the necks of the flasks were covered with Parafilm, and each flask was then covered in tin foil and left at room temperature. Samples of soil were taken from each microcosm at weeks 1, 2, 4, and 8. Prior to removal of soil from the microcosm, the soil was mixed well in order that there was no bias in the sampling procedure. A sample of soil of around 15 g was removed from each microcosm. Subsamples of the homogenized soil removed were extracted immediately and kept at 4 °C. To 1 g of this moist soil, 10 mL of deionized water was added. The tubes were placed in a sonication bath for 30 min. A 6-mL sample of the supernatant fluid was removed after centrifugation from each tube and kept in cold storage. Commercially Available Fertilizers. Kerry Enhancer MG 95 (Kerry Algae Ltd., Tralee, Ireland) is a black granular solid to be applied at the rate of 700-800 g/ha for vegetables or 400 g/500 m2 for the golf courses. It is produced by hightemperature extraction and alkaline hydrolysis of Ascophyllum nodosum, Laminaria spp., Fucus spp., and Chondrus crispus. The other ferilizer (SM6, from Chase Organics GB Ltd., Surrey, England) is a seaweed extract that has a solid seaweed amount of 30% (w/v). It represents fertilizers that are produced by a more gentle process (water extraction of A. nodosum, Laminaria spp., and Fucus serratus at low temperature). The annual application rate is dependent on the crops and can vary between 1 and 11 L of SM6/ha. Field Survey of Seaweed Manured Soils. (a) SeaweedFertilized Soil. Inverewe Garden, northwest Scotland (Figure 1A), had been created with the aid of extensive fertilization with seaweed. Samples from the walled garden (Figure 1D), where the seaweed was applied (site 1), and non-seaweed fertilized tracts of the garden were sampled during January 2002 (sites 2 and 3). Within the walled garden, soils were sampled on a grid intercept, with 8 transects of 100 m through the beds, sampling every 10 m. Outside the walled garden, samples were collected less intensively along linear transects at sites 2 and 3. Sampling of machair soils on the island of South Uist, Outer Hebridies, Scotland (Figure 1B), was conducted in late August 2001. These sites have been fertilized with seaweed 952
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 5, 2003
FIGURE 1. Map of South Uist and Inverewe Garden on the west coast of Scotland. Panel B shows sampling locations on South Uist, while panel C shows sites 2-4 in more detail. Panel D shows the walled garden at Inverewe, outlined by a dashed line, with the location of control sampling points (2 and 3). from prehistoric times (1, 2). Since the 1950s, this traditional agriculture was abandoned with modern fertilizers being used. Over the last 5 yr, there has been a reversion to traditional seaweed manuring supported by government grants. The sampling sites are given in Figure 1B. A detailed survey of soil and crops (barley, rye, oat, and potato) was conducted around sites 2-4. Sites 1, 5, 7, and 8 were on the edge of the machair terrace from which cultivated machair soils (25 m from the edge of the terrace), sand at the base of the terrace, beach sand, and beach sand from under piles of decomposing seaweed were collected; for the soils/sands, the top 10 cm was sampled. Site 2 (which is on an island, Orasaigh, connected to the main island by a sand causeway) and sites 4, 6, and 9 were peats. Inland sites 4, 6, and 9 were selected as controls; they had no previous cultivation. Orasaigh was used for potato cultivation, but this and all other cultivation ceased before the use of chemical fertilizers and is only used currently to graze sheep. Where present crops were taken for sampling, only one sample of each substrate was taken at each sampling location except for sites 2-4 where (at least) triplicate samples were obtained. Each sample was collected at 2.5-m intervals along a linear transect perpendicular to the coastline. For site 3, the croft had samples taken along a transect through the field (the field systems are long and narrow, running from the main road to the coast). The soils gradate along the transect shown in Figure 1C and range from sands at the coasts through to peats. All the field systems, from the coast to the croft, are fertilized with seaweed. At the croft building, abandoned vegetable plots (all peats), which had not received chemical fertilizers, were also sampled. Analytical Procedures. (a) Chemicals and Reagents. Dimethylarsinic acid (DMA(V)) was obtained from Sigma Chemicals, and methylarsonic acid (MA(V)) was a gift of Prof. W. R. Cullen (Vancouver, Canada). Sodium arsenate and sodium arsenite, both reagent grade, were supplied by Merck.
FIGURE 2. Structures and used names for four common arsenosugars. Other arsenic standards such as dimethylarsinoyl ethanol (DMAE) and four different arsenosugars (structures in Figure 2) [sugar-OH (synthesized), sugar-PO4, sugar-SO3, and sugarSO4 (extracted from seaweed)] were gifts from Dr. W. Goessler (Karl-Franzens University, Graz, Austria). The standard reference material IAEA 140 (common Fucus spp., IAEA, Monaco) was used for the total arsenic determination and arsenic speciation analysis, while for the survey the soil certified reference material GBW07406 (Institute of Nuclear Research Academia Sinica, Shanghai, China) was used. (b) Seaweed Extraction Procedure for Species Analysis. A total of 0.1 g of dried seaweed, fertilizer, or reference material was mixed with 10 mL of methanol/water (1:10 (v/ v)) and sonicated for 10 min. The supernatant was removed after centrifugation. This procedure was repeated three times. The supernatants of each extraction were combined and evaporated to 1 mL under reduced pressure at room temperature using a rotor evaporator. The evaporated extracts were stored in a cool room until analysis and diluted prior to the separation with water. (c) Total Arsenic Analysis. A 0.25-g sample from each of the oven-dried soil or seaweed samples was weighed into digestion tubes, and 2.5 mL of 69% nitric acid (Analgar grade) was added and left overnight. Afterwards 2.5 mL of H2O2 was added and left for 30 min at 50 °C, followed by boiling for 3 h under reflux until the solution became clear. After being cooled, the solution was centrifuged, and the supernatant was diluted with deionized water to 10 mL. A 1 mL of this sample was then diluted to 10 mL with a solution of 10% HCl, 10% KI, and 5% ascorbic acid. To determine arsenic levels in soil solution extracts, 1 mL of the sample was diluted to 10 mL, again with the HCl/KI/ascorbic acid mixture. Appropriate quality control, blanks and spiked samples, and a certified soil reference material GBW07406 were included for every 40 samples. After left stand overnight to ensure complete reduction, the digest was analyzed for arsenic using hydride generation-atomic absorption spectrometry using a Perkin-Elmer (U.K.) FIAS 100 flow injection hydride generator interfaced with a Perkin-Elmer AAnalyst 300 atomic absorption spectrometer. Quality control for the soil analysis of total arsenic showed that the analytical procedures were appropriate. Instrumental limits of detection were 0.2 ng mL-1 based on 3 times standard deviation of the background noise. Recovery of the soil reference material GBW07406 was 87.6 ( 0.7% (n ) 3) of its certified value of 220 ( 7 mg kg-1. Recovery of the arsenic spike was 90.4% (n ) 3). (d) Analytical Method for Arsenic Speciation in Water Extract. The methanol/water extracts were filtered (0.45-µm nylon membrane, Supelco, Bellefonte, PA) and directly injected (20 µL loop) onto an anion-exchange column (PRP X-100, 250 × 4.6 mm). As the mobile phase, a 30 mM H3PO4 solution adjusted to pH 6.0 with NH3 and an addition of 100 µg L-1 Cs as a continuous internal standard were used and coupled to an ICP-MS (Spectromass 2000, Spectro Analytical Instruments) via a Meinhard nebulizer and a water-jacked cyclonic spray chamber. The MS was set on m/z 75, 77, and 133 with a dwell time of 500 ms. Arsenic was monitored on m/z 75, whereas m/z 77 was used to check for the occurrence
FIGURE 3. Chromatograms of eight arsenic standards determined by anion-exchange liquid chromatography coupled to ICP-MS (m/z 75 monitoring) using 30 mM ammonium phosphate buffer, pH 6.0. 1, arsenite, 30 ng mL-1; 2, sugar-OH; 3, DMA(V), 100 ng mL-1; 4, MA(V), 50 ng mL-1; 5, sugar-PO4; 6, arsenate, 50 ng mL-1; 7, sugarSO3; 8, sugar-SO4. of 40Ar37Cl+ because this cluster would be formed with the same likelihood as 40Ar35Cl+, and this would give an interference on m/z 75. Cs was measured on m/z 133 for monitoring the plasma stability throughout the chromatographic runs. Arsenic species solutions of 10, 20, 50, and 100 µg of As/L of arsenite, DMA(V), MA(V), and As(V) were made fresh daily and used for quantification. These standards were not only used to quantify these four species but also for the arsenosugars because of the limited amount of arsenosugar standards available since the response of all four arsenic species is very similar; therefore, we assume that these responses can be applied to the arsenosugars as well. A chromatogram of the standard species is shown in Figure 3. Eight different arsenic species can be separated and detected in the lower nanograms per milliliter range. Estimated detection limits based on the 3 times standard deviation of 10 ng mL-1 DMA(V) solution is 1 ng mL-1, whereas the detection limits for later-eluting peaks (>400 s) are considerably higher. For example, the concentration of sugar-SO4 in L. digitata of 0.1 mg kg-1 can just be detected. Samples in which arsenic species elute in the void volume were checked by using a cation-exchange method (Supelcosil with 20 mM pyridine buffer, pH 3; described in ref 10) for the occurrence of dimethylarsinoyl ethanol (DMAE), arsenobetaine, or arsenocholine.
Results and Discussions Arsenic Speciation in Seaweeds and Seaweed Fertilizers. The seaweed used contained a relatively high arsenic concentration of about 24 mg kg-1 dry mass and 74 mg kg-1 for F. vesiculosus and L. digitata, respectively (Table 1). These results are comparable with those published before (10, 11). These concentrations are higher than the permissible level of arsenic in agricultural soil (50 mg kg-1) (12). If an application rate of 25 t of seaweed/ha is assumed, it can be expected that the arsenic concentration in the soil would increase by more than 1 mg kg-1 for each application. In Table 1, the quantitative data from two different seaweedbased fertilizers are shown. The arsenic concentration in the solid product (Kerry Algae) and the suspension (SM6) are of the same order of magnitude as the dry mass of the used seaweed between 13 and 14 mg kg-1. If the application rates of 11 L of SM6/ha (recommended for potato) is followed, the arsenic addition would be negligible from one application. A mild methanol/water extraction of seaweed seemed to be reasonably efficient with extraction efficiencies of about 70% (Table 1), which is in agreement with previous studies (10, 13). Although no appropriate standard reference material with certified arsenic species was available, IAEA 140 (Fucus VOL. 37, NO. 5, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
953
TABLE 1. Arsenic Species in Fucus vesiculosus and Laminaria digitata and Two Commercially Available Seaweed-Based Fertilizers As Compared to Standard Reference Material Fucus sp., IAEA 140 (Certified Value: 42.2-46.4 mg/kg Total Arsenic)a
seaweed
total arsenic
F. vesiculosus L. digitatab Kerry Algae Enhancer SM6 IAEA 140
25 ( 4 74 ( 2 41.1 ( 2.0 13.4 ( 1.0 40.4 ( 2.2
extraction efficiency (%) As(III) 73 69 100 68c 82
bd bd bd bd bd
DMA(V)
MA(V)
As(V)
sugar-OH
sugar-PO4
sugar-SO3
sugar-SO4
0.01 0.87 26.8 ( 1.9 1.3 ( 0.4 3.7 ( 0.5
bd bd bd bd bd
0.04 0.38 1.2 ( 0.6 0.27 ( 0.01 1.1 ( 0.5
0.69 1.86 8.9 ( 1.1 2.3 ( 0.2 9.9 ( 0.4
0.9 5.3 1.9 ( 0.8 0.43 ( 0.12 0.8 ( 0.4
10 42 4.3 ( 0.8 1.5 ( 0.1 9.6 ( 0.3
6.1 0.1 bd 4.7 ( 0.3 11.6 ( 0.4
a All values are given in mg/kg dry mass and the standard error of the mean for 3 replicates. Recovery from the chromatographic column is quantitative (sum of LC-ICP-MS is 100 ( 3% of the totals arsenic in the extract). bd, below detection limits of 0.01 mg/kg. b Unknown species: 0.5%. c Only the filtered liquid was measured which accounts for 68.4%.
TABLE 2. Arsenic Concentration in Environmental Materials Collected from a Machair Agronomic System Fertilized with Seaweeda location
total As (mg kg-1 dw)
Orasaigh Island Survey beach sand 1, 3, 7, 8 sand under seaweed 1, 3, 7, 8 edge of terrace 1, 3, 7, 8 sand under crops 1, 3, 7, 8 unfertilized peat 4, 5, 6 seaweed-fertilized peat (Orsaigh) 2
0.71 ( 0.14 0.60 ( 0.13 0.50 ( 0.10 0.71 ( 0.23 0.64 ( 0.11 4.34 ( 1.09
Croft at Sites 3 and 4 beach sand 3 sand under seaweed 3 edge of terrace 3 sand under crops 1 2, 3 sand crops 2 2, 3 peat crops 3 peat (abandoned croft garden) 3 uncultivated peat 4 stormcast seaweed mixture mainly Laminaria spp.
0.65 ( 0.05 0.57 ( 0.30 0.28 ( 0.01 1.30 ( 0.20 0.56 ( 0.06 1.98 ( 0.38 1.48 ( 0.22 0.35 ( 0.02 53 ( 13
description
FIGURE 4. Chromatogram of the methanol/water extracts of the seaweed used as fertilizer for soil. (s) Fucus vesiculosus; (- -) Laminaria digitata; 2, sugar-OH; 3, DMA(V); 5, sugar-PO4; 6, arsenate; 7, sugar-SO3; 8, sugar-SO4. sp.) was used to demonstrate the quality of the speciation data (Table 1). More than 90% of the extractable arsenic is specified to be in the form of arsenosugars shown in Figure 2, while inorganic arsenic is less than 5% in IAEA 140. In the seaweed extracts, the arsenic species were identified by comparing the retention times with those of standards using the anion-exchange chromatography (Figures 3 and 4). The majority of the extractable arsenic species were in the form of arsenosugars. L. digitata contained mainly sugar-SO3 and sugar-PO4, while F. vesiculosus contained mainly sugar-SO3 and sugar-SO4 in addition to sugar-PO4. Only a small amount of sugar-OH, DMA(V), and arsenate (As(V) has been identified (Table 1). In commercially available seaweed fertilizer (SM6), which is made by a gentle low-temperature water extraction, mostly arsenosugars have been determined. When an alkaline extraction process was performed (Kerry Enhancer) to generate a seaweed-based fertilizer, the arsenosugars decompose mainly to DMA(V), whereas a mild acidic digestion would transfer the arsenosugars only to their unsubstituted arsenosugar in which the side chain is replaced by -OH (14). Although the application rate for those fertilizers is low with a maximum of 11 L/ha, it has to be shown in microcosm experiments whether a significant arsenic uptake into the crops takes place. Long-Term Fertilization Study. A mixture of storm-cast seaweed mainly Laminaria spp. (L. digitata, L. hyperborea, L. saccharina) were analyzed and gave an average arsenic concentration of 53 ( 13 mg kg-1. The variability of results is expected because of the variability of the different seaweed species (15) since the storm-cast bundles of seaweed consist of a variety of seaweed species. Levels of arsenic in sands and peats throughout Orasaigh Island (site 2, Figure 1C) were generally low, with average values ranging from 0.5 to 0.7 mg kg-1 (Table 2). The arable soils, in contrast, would have 954
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 5, 2003
Crops rye oats barley potatoes
2, 3, 4 2, 3, 4 2, 3, 4 2, 3, 4
0.09 ( 0.01 0.07 ( 0.01 0.11 ( 0.03 0.11 ( 0.01
a Site locations are given in Table 1. Arsenic levels are the averages of at least 3 replicates including standard error.
seaweed added to them constantly, season to season, and might have been expected to have elevated arsenic if the seaweed-derived arsenic remained in the soils. The fact that elevated arsenic in the arable soils (4.34 mg kg-1) were not as high as estimated, on the basis of a 25 t ha-1 yr-1 application rate, suggests that most of the seaweed-derived arsenic has been leached from the soil profile since the application almost a century ago. This analysis is based on average levels for the island. The data presented in Table 2 for sites 2-4 (Figure 1C) are located by a rich source of seaweed that collects between the Island of Orasaigh and the mainland, which are connected by a causeway. The arable machair sands located right by the coast are elevated in arsenic (1.3 mg kg-1) as compared to the neighboring beach sand and sand under the arable terrace (0.28 mg kg-1), significant at the p < 0.001 level (oneway analysis of variance). Arsenic levels drop in machair sands further from the coast, only to rise again as the sand gradates into peats with levels on these seaweed-manured peat grasslands rising to 1.98 mg kg-1. Figure 1C compares to an arsenic level of 0.35 mg kg-1 in a neighboring nonagronomic peat (site 4). So peat arsenic levels have been elevated 6-fold by seaweed manuring in these grasslands.
TABLE 3. Soil Arsenic Levels at Inverewe Garden, Scotlanda description
site
arsenic (µg/g) ( SE
no. of samples
walled gardens outside the walled gardens outside the walled gardens soil from Inverewe no treatment 6 months before sampling
1 2 3
12.6 ( 0.6 4.07 ( 0.86 3.57 ( 0.69
86 16 14
1
6
treatment with seaweed prior to sampling
1
17.7 ( 2.4 (total As) 1.27 ( 0.96 (soluble As) 22.2 ( 5.8 (total As) 1.45 ( 0.90 (soluble As)
a
TABLE 4. Soluble Arsenic Concentration in Soil Taken from Microcosmsa soluble As (µg g-1) week
soil + L. digitata
soil + F. vesiculosus
1 2 4 8
1.26 ( 0.21 0.64 ( 0.06 0.55 ( 0.15 1.15 ( 0.67
0.23 ( 0.06 0.02 ( 0.01 0.28 ( 0.08 0.44 ( 0.14
a An application rate of 50 g of seaweed kg-1 soil, n ) 4. The standard error represents four replicates. Control soil: 1.57 ( 0.33 µg g-1 as total As,
