Arsenic Speciation in Plankton Organisms from Contaminated Lakes

ARTICLE pubs.acs.org/est

Arsenic Speciation in Plankton Organisms from Contaminated Lakes: Transformations at the Base of the Freshwater Food Chain Guilhem Caumette, Iris Koch, Esteban Estrada, and Ken J. Reimer* Environmental Sciences Group, Royal Military College of Canada, P.O. Box 17000 Station Forces, Kingston, Ontario K7K 7B4, Canada ABSTRACT: The two complementary techniques high performance liquid chromatography-inductively coupled plasma-mass spectrometry (HPLC-ICP-MS) and X-ray absorption near edge structure (XANES) analysis were used to assess arsenic speciation in freshwater phytoplankton and zooplankton collected from arsenic-contaminated lakes in Yellowknife (Northwest Territories, Canada). Arsenic concentrations in lake water ranged from 7 μg L
1 in a noncontaminated lake to 250 μg L
1 in mine-contaminated lakes, which resulted in arsenic concentrations ranging from 7 to 340 mg kg
1 d.w. in zooplankton organisms (Cyclops sp.) and from 154 to 894 mg kg
1 d.w. in phytoplankton. The main arsenic compounds identified by HPLC-ICP-MS in all plankton were inorganic arsenic (from 38% to 98% of total arsenic). No other arsenic compounds were found in phytoplankton, but zooplankton organisms showed the presence of organoarsenic compounds, the most common being the sulfate arsenosugar, up to 47% of total arsenic, with traces of phosphate sugar, glycerol sugar, methylarsonate (MMA), and dimethylarsinate (DMA). In the uncontaminated Grace Lake, zooplankton also contained arsenobetaine (AB). XANES characterization of arsenic in the whole plankton samples showed AsV
O as the only arsenic compound in phytoplankton, and AsIII
S and AsV
O compounds as the two major inorganic arsenic species in zooplankton. The proportion of organoarsenicals and inorganic arsenic in zooplankton depends upon the arsenic concentration in lakes and shows the impact of arsenic contamination: zooplankton from uncontaminated lake has higher proportions of organoarsenic compounds and contains arsenobetaine, while zooplankton from contaminated area contains mostly inorganic arsenic.

’ INTRODUCTION Arsenic cycling in the environment and through the aquatic food chain is crucial in understanding the origin of arsenic species with ranging toxicities, especially the nontoxic arsenobetaine compound. Arsenic in the marine environment has been widely studied and shows evidence of arsenobetaine being acquired through the food chain,1
3 as a possible osmolyte for saline environments.4 Marine zooplankton contain arsenobetaine as a major compound,5,6 while marine phytoplankton contain mainly arsenosugars.7
11 There are several theories regarding arsenobetaine formation in the marine environment, but it is most likely formed at the base of the food web.12 In the freshwater environment, osmolytic requirements are not the same, but arsenobetaine has been measured in freshwater fish and water bugs.13,14 Information on arsenic speciation in freshwater plankton is very scarce. Some studies on arsenic speciation in freshwater algae from an arsenic-contaminated river13,15 showed the presence of methylated arsenic and arsenosugars. The culture and incubation of the microalgea Chlorella sp. and Monoraphidium arcuatum in an arsenic-controlled environment16 did not show methylation of inorganic arsenic by these organisms, with the only effect measured being the reduction of AsV to AsIII. On the other hand, Murray et al.17 showed the ability of the microalgae Chlorella vulgaris cultured in a controlled environment to methylate inorganic arsenic to MMA and DMA, and to form arsenosugars. The same effect was observed by Miyashita et al.18 with Published 2011 by the American Chemical Society

the green microalga Chlamydomonas reinhardtii. Thus, the ability of freshwater phytoplankton to form organoarsenic compounds remains inconclusive. Moreover, no information has been reported so far on arsenic speciation in lake or river zooplankton. The only work on freshwater zooplankton to our knowledge concerns a culture of Daphnia magna exposed to inorganic arsenic.19 The arsenic species identified in Daphnia were inorganic arsenic AsIII and AsV with traces of MMA and DMA. This work reports the investigation on arsenic speciation in freshwater phytoplankton and freshwater zooplankton from arsenic-contaminated lakes (Yellowknife, NT, Canada) to elucidate arsenic transformation pathways in the freshwater environment. The hyphenated technique HPLC-ICP-MS is typically used for arsenic speciation analysis of sample extracts. Because mild extraction methods used for HPLC often extract less than 50% of the total arsenic from a contaminated biological samples, 20 X-ray absorption spectroscopy (XAS) was used as a complementary technique to assess the speciation of the nonextractable arsenic. This technique has often been used for arsenic speciation in biological organisms,21
23 because whole solid or frozen samples can be analyzed, therefore giving information on the Received: July 20, 2011 Accepted: October 23, 2011 Revised: October 14, 2011 Published: October 23, 2011 9917

dx.doi.org/10.1021/es2025092 | Environ. Sci. Technol. 2011, 45, 9917–9923

Environmental Science & Technology

Figure 1. Map of Yellowknife and the sampling locations. (1), (2), and (3) are Kam lake sampling locations with a gradient of contamination from point (1). (4) is located on Long Lake, and (5) is on Grace Lake.

speciation of 100% of the element analyzed, with no need for extraction or sample preparation. XAS was used on whole plankton samples as well as residues from water extraction. Because the detection limits are much higher than HPLC-ICP-MS methods, the use of both methods is a powerful combination to completely characterize arsenic in environmental samples.

’ MATERIALS AND METHODS Sampling. Three lakes of interest with a varying arsenic contamination were chosen for sample collection. They were Grace Lake (62°250 1100 N, 114°260 1400 W), Kam Lake (62°250 3700 N, 114°230 5000 W), and Long Lake (62°280 4100 N, 114°270 3300 W) from Yellowknife, NWT. The Con Mine gold mine, situated northeast of Kam Lake (Figure 1), historically contaminated Kam Lake. Sediments, pore water (interstitial water in sediment), lake water, phytoplankton, and zooplankton samples were collected from the sampling points shown in Figure 1. Lake parameters such as pH, temperature, dissolved oxygen, and conductivity were monitored. Each location was sampled during the first, second, and third weeks of June 2010 to target the peak of phytoplankton production.24 Sediments were collected with a Ponar-Grab (Wildco Standard 9  9 Ponar Grab). Sediments and pore water were separated by centrifugation (15 min at 2000 rpm) within 3 h of collection and stored frozen until analysis. Phytoplankton was collected using a Van-Dorn sampler (Wildlife Supply Company, Horizontal Van Dorn Sampler) in the water column (1
3 m deep) and isolated as follows: zooplankton was first removed from phytoplankton using a 45 μm sieve, and then phytoplankton was retained by filtration on a 0.45 μm filter (nylon, 0.45 μm pore size, 47 mm diameter, Millipore Corp., Billerica, MA). The water and filters with phytoplankton were kept frozen until analysis. Zooplankton was collected using a zooplankton net (Wisconsin Net, 153 μm pore size). Zooplankton species were identified immediately after collection by using a stereoscopic microscope. Zooplankton species were separated by size using sieves of 250, 180, and 45 μm pore size. The three groups of zooplankton were

ARTICLE

labeled ø250 (organisms larger than 250 μm and retained on the 250 μm pore size sieve), ø180 (organisms ranging from 180 to 250 μm and retained on the 180 μm pore size sieve), and ø45 (organisms ranging from 45 to 180 μm and retained on the 45 μm pore size sieve). Phytoplankton and zooplankton samples were kept frozen at
80 °C until freeze-drying and analysis. Acid Digestion. For acid digestion, 70% nitric acid (reagent grade, Fisher, and Optima grade, Fisher), 30% hydrogen peroxide (in water, reagent grade, Caledon Laboratory), and deionized water (18 MΩ) were used. Whole samples and residues of extraction were digested for total analysis. For biological tissues, approximately 0.1 g of dry sample was weighed in a test tube, and 10 mL of nitric acid (reagent grade) was added. The test tube was heated at 140 °C until the sample had evaporated to near dryness. Next, 2 mL of hydrogen peroxide was added and evaporated to near dryness at 140 °C. The residual sample was then diluted in 10 mL of 2% nitric acid (Optima grade, Fisher) in deionized water for ICP-MS analysis. For sediment analysis, approximately 0.1 g of dry sample was weighed in a test tube, 9 mL of nitric acid (reagent grade) and 3 mL of hydrochloric acid (reagent grade, Fisher) were added, and the test tube was heated at 140 °C for evaporation to near dryness. A further 2 mL of nitric acid (reagent grade) was added and evaporated in the same way. The residual sample was then diluted in 10 mL of 2% nitric acid (Optima grade, Fisher) in deionized water for ICP-MS analysis. Water Extraction. Zooplankton and phytoplankton samples were extracted with water after freeze-drying. Approximately 0.1 g of dry sample was weighed in a test tube, and 10 mL of deionized water was added. The extraction method was optimized for extraction efficiency of arsenic compounds, using the CRM Sea Lettuce (BCR-279, Institute for Reference Materials and Measurements, Belgium) and DORM-2 (Dogfish Muscle CRM, National Research Council of Canada). Increasing the temperature of the extracting water showed a higher efficiency of extraction. Extraction at room temperature gave 93% of total arsenic extracted for DORM-2 while only 37% for Sea Lettuce. At 40 °C, the Sea Lettuce extraction efficiency was 41%, and a maximum of 53% was obtained at 60 °C. DORM-2 extraction efficiency remained at 93% at both temperatures. The sonication time and centrifugation step did not significantly affect the extraction efficiency. The optimized extraction conditions were therefore a water extraction at 60 °C with 4 h shaking, followed by 10 min sonication and 15 min centrifugation at 4000 rpm. These conditions were used for all phytoplankton and zooplankton samples. ICP-MS for Total Arsenic Analysis. All digested samples were diluted with 2% nitric acid (Optima grade, Fisher) to ensure that arsenic concentrations were within the calibration limits. Calibration standards of 1, 5, 10, 25, 50, 100, 250, and 500 μg kg
1 of arsenic as AsV (arsenic pentoxide, Inorganic Ventures, 1000 μg mL
1) were used for total arsenic analysis. The instrument was an ICPMS DRC II from Perkin-Elmer (Perkin-Elmer, MA). The nebulizer was concentric-type, the flow of argon was 0.87 L/ min, the RF power was 1300 W, and the lens voltage was 6 V. The analysis mode used was peak hopping, the dwell time was 100 ms, and the instrument read 10 sweeps and three replicates per sample. Internal standards were rhodium and indium (1000 μg mL
1, SCP Science). The masses monitored were 75As with 115In and 103Rh for signal stability. For quality control, two standard solutions of 5 and 50 μg L
1 arsenic were analyzed every 10 samples, and results were accepted only if recovery was in an acceptable range (e.g., 90
110%). The deviation of the instrument 9918

dx.doi.org/10.1021/es2025092 |Environ. Sci. Technol. 2011, 45, 9917–9923

Environmental Science & Technology was below 1.3% during the analysis, and the limit of detection of the instrument was calculated as 0.5 μg/L. Total arsenic was measured in three certified reference materials (CRM): marine sediments (MESS-3), dogfish tissue (DORM-2), and Sea Lettuce (BCR-279). These three CRMs were used to represent as closely as possible sediments, zooplankton, and phytoplankton tissues. Certified values are 18 ( 1.1 mg kg
1 arsenic for DORM-2, 3.1 ( 0.2 mg kg
1 arsenic for Sea Lettuce, and 21.2 ( 1.1 mg kg
1 arsenic for MESS-3. The measured concentrations on three extractions of each CRM (n = 3) were found, respectively, as 18.8 ( 0.3, 3.4 ( 0.1 mg kg
1, and 22.1 ( 0.2 mg kg
1. These values are within the standard deviation of the certified material, except for the Sea Lettuce where the concentrations found were 10% higher than the certified, and this was considered to be sufficiently close that the results were deemed acceptable. HPLC-ICP-MS for Arsenic Speciation Analysis. Water extracts were analyzed by HPLC-ICP-MS for arsenic speciation, with anion and cation exchange columns. The chromatographic system was a Perkin-Elmer pump (Flexar LC pump, PerkinElmer, MA). The anion exchange column was a Hamilton PRPX100, 4.6  150 mm, 10 μm, with matching guard column. Anion mobile phases were (A) 4 mM ammonium nitrate (99.999% purity, Aldrich) and (B) 60 mM ammonium nitrate in deionized water, both adjusted to pH 8.7. The gradient of elution, developed by Watts et al. (Watts et al., 2008), was: 100% A 0
2 min; 100% B 3
6.5 min; 100% A 7.5
10.75 min; 100% B 11
13 min; 100% A 13.25
15 min with a flow rate of 1 mL min
1. The cation exchange column was a Chrompack Ionosphere C, 3  100 mm or 3  150 mm, 5 μm, with matching guard column. The mobile phase used was 20 mM pyridine (99% purity, Sigma), pH 2.7, 1.5 mL min
1. Standard mixtures of dimethylarsinate (DMA, cacodylic acid, 98%, Fluka), arsenobetaine (AB, 95%, Wako), trimethylarsine oxide (TMAO, 95%, Wako), arsenocholine (AC, arsenocholine bromide, 95%, Wako), and tetramethylarsonium (Tetra, tetramethylarsonium iodide, 95%, Wako) were used for cation exchange (5, 10, 25, 50, 100, and 250 μg kg
1). AB, arsenite (AsIII, trioxide arsenic, Inorganic Ventures, 1000 μg mL
1), DMA, monomethylarsonate (MMA monosodium acid methane arsonate sesquihydrate, 99%, Chem Service), and arsenate (AsV arsenic pentoxide, Inorganic Ventures, 1000 μg mL
1) were used for anion exchange (5, 10, 25, 50, 100, and 250 μg kg
1). Arsenosugars used as standards were extracted from brown algae (Fucus vesiculosus) collected in Nova Scotia, Canada. The extraction method of arsenosugars followed the steps described by Madson et al.25 Arsenosugars were identified by matching retention times, and arsenosugar concentrations were calculated by using the curve calibration of the closest arsenic compound standard analyzed. The software PeakFit (Systat Software Inc. 2008) was used to measure the concentration of each compound identified. The concentrations of DMA, phosphate sugar (sugar 2), MMA, sulfonate sugar (sugar 3), AsV, and sulfate sugar (sugar 4) were measured from separation on the anion exchange column, and the concentrations of glycerol sugar (sugar 1) and AB were measured with the cation exchange column. DMA concentration was measured with the anion exchange column only, as DMA overlaps with anion compounds on the cation exchange column. In the same way, AB was measured with cation exchange column where it is well separated from the other compounds (Figure 2). The concentration of AsIII was calculated by subtracting the concentration of sugar 1 from the second peak of the anion

ARTICLE

exchange chromatogram (Figure 2). The average column recovery was calculated as 89% for anion and 83% for cation exchange columns. The limit of detection of the instrument was, respectively, 1, 0.6, 0.5, 0.5, and 0.6 μg kg
1 for AB, AsIII, DMA, MMA, and AsV with anion exchange column and 0.2, 0.2, 0.9, 4, 11 μg kg
1, respectively, for DMA, AB, TMAO, AC, and Tetra with cation exchange columns. For each sample, two replicates were analyzed by HPLC-ICP MS: one water extract and one water extract spiked with 50 μg g
1 standards of AB, AsIII, DMA, MMA, and AsV. The spikes of standard were used to identify the peaks (Figure 2). Each sample was run in both anion and cation columns. Typical chromatograms obtained for Grace Lake zooplankton’s extract and standards of arsenic compounds are presented in Figure 2. For quality control, the two CRMs DORM-2 and sea lettuce were analyzed and quantified using HPLC. Total arsenic concentration in CRMs was calculated by adding total concentration measured in the water extract and total concentration measured in the residues. Total arsenic concentration in Sea Lettuce was found at 3.5 mg kg
1, which was within 13% of the certified value of 3.1 ( 0.2 mg kg
1. The extracted solution from the Sea Lettuce was measured at 1.8 mg kg
1 with 1.7 mg kg
1 in the residue, which gives an extraction efficiency of 52%. Extracted solution contained 0.7 mg kg
1 of AsV, 0.2 mg kg
1 of AB, 0.2 mg kg
1 of sugar 1, 0.2 mg kg
1 of sugar 3, 0.1 mg kg
1 of sugar 2, and traces of Tetra, which gives a total of 1.4 mg kg
1 arsenic and a column recovery of 87%. Total arsenic in DORM2 was found as 20.7 mg kg
1, within 15% of the certified value of 18 ( 1.1 mg kg
1. The extracted solution from DORM-2 was measured at 18.7 mg kg
1 with 2 mg kg
1 in the residue, which gives an extraction efficiency of 93%. The extracted arsenic from DORM-2 contained 18 mg kg
1 of AB and 0.3 mg kg
1 of Tetra; certified values are 16.4 ( 1.1 mg kg
1 AB and 0.25 ( 0.05 mg kg
1 Tetra. The measured AB concentration was slightly outside the range of the certified concentration but still within 15% of the certified value, with 96% column recovery for this sample. XANES Analysis. XANES spectra were collected at the Pacific Northwest Consortium X-ray Science Division (PNC/XSD) facilities, Sector 20 at the Advanced Photon Source (APS), Argonne National Laboratory, using methods similar to those described in Smith et al.21 Freeze-dried zooplankton samples and residues from water extraction were ground into powder and packed into vials for analysis. Filters containing phytoplankton were dried, rolled in length, and fixed on the sample holder. All samples were placed between two layers of Kapton tape. XANES spectra of the arsenic K-edge (11 868 eV) were fit within
20 to +30 eV to E0 using Athena software. The Si(111) double-crystal monochromators was calibrated using the first inflection point of the gold LIII absorption edge (11 919.7 eV). A reference gold foil was measured simultaneously with samples. Frozen AsIII and AsV, liquid As(glutathione)3, arsenosugars, DMA, and MMA standards measured previously by our group21 were used for fitting. The detection limits were about 1 mg kg
1.

’ RESULTS AND DISCUSSION Description and Conditions of the Lakes Analyzed. The full description of the lakes studied is presented in Table 1. Kam Lake, Long Lake, and Grace Lake were 4.1, 2.9, and 3 km long, respectively. Kam Lake was chosen because it is directly contaminated by Con Mine effluent and tailings. Three different 9919

dx.doi.org/10.1021/es2025092 |Environ. Sci. Technol. 2011, 45, 9917–9923

Environmental Science & Technology

ARTICLE

Figure 2. HPLC-ICP-MS profiles of 75As in standards and zooplankton extracts from Grace Lake using (a) anion exchange HPLC and (b) cation exchange HPLC. Zooplankton samples have been separated into two groups of size: 250 μm organisms and 180 μm organisms.

Table 1. Description and Characteristics of the Lakes Studieda lake

temp (°C)

pH

DO (mg/L)

SpC (mS/cm)

[As] sediments (mg kg
1)

[As] pore water (μg L
1)

[As] water (μg L
1)

Kam(1)

15.0 ( 0.9

7.7 ( 0.3

11.5 ( 0.2

0.31 ( 0.01

698 ( 2

779 ( 33

250 ( 100

Kam(2)

15.0 ( 0.3

7.7 ( 0.4

11.2 ( 0.2

0.31 ( 0.01

223 ( 14

232 ( 27

158 ( 42

Kam(3)

15.0 ( 1

7.6 ( 0.3

11.5 ( 0.1

0.31 ( 0.01

123 ( 8

281 ( 10

145 ( 66

Long

16.3 ( 0.1

7.9 ( 0.2

11.5 ( 0.1

0.30 ( 0.01

141 ( 10

290 ( 9

51 ( 3

Grace

15.8 ( 0.5

7.6 ( 0.5

11.7 ( 0.4

0.14 ( 0.01

34 ( 5

64 ( 40

7(2

a

Data are given as an average for the three sampling weeks, with the corresponding standard deviation (n = 3). Kam(1) is the closest to historical mine discharge, and Kam(3) is furthest. Neither Long nor Grace lakes were directly affected by mining activities, and thus conditions are assumed to result from the natural geology of the location.

positions were sampled in this lake: at the source of contamination (position 1 in Figure 1), in the middle of the lake (position 2), and as far as possible from the contamination source (position 3). Grace Lake and Long Lake were chosen as background lakes, in other words, not directly impacted by mine effluent or tailings. These lakes are far enough from the mine to avoid direct contamination, and Grace Lake drains into the Kam Lake water system. However, the naturally elevated arsenic concentration in the Yellowknife area (Yellowknife bedrock contains arsenopyrite bound to gold) makes these lakes not arsenic-free.26 The pH, temperature, and dissolved oxygen conditions were similar in the three lakes analyzed, with the only difference noted being the conductivity, which was lower in Grace Lake than in the other lakes.

Arsenic concentrations were measured in the three samples from the three sampling weeks to obtain a meaningful average on the concentration. The standard deviations obtained are presented in Table 1. Arsenic measurements in sediments and pore water were very repeatable over the 3 weeks with a standard deviation below 15% for all samples analyzed. Arsenic measurement in water showed a wider variation, which might be due to the mobility of water in the lake. The arsenic concentration obtained in sediments ranges from 34 mg kg
1 in uncontaminated Grace Lake, up to 698 mg kg
1 in the highly contaminated area of Kam Lake. Arsenic in pore water and water was 1000 times lower, ranging from 64 to 779 μg L
1 in pore water and from 7 to 250 μg L
1 in water. The values measured are the same order as those measured previously by our group.27 The results 9920

dx.doi.org/10.1021/es2025092 |Environ. Sci. Technol. 2011, 45, 9917–9923

Environmental Science & Technology

ARTICLE

Table 2. HPLC-ICP-MS Data of Plankton Samples: Total Arsenic Concentration, Arsenic Compounds Identified, and Extraction Efficiencya arsenic compounds (% of total arsenic extracted) sample


1

V

total As (mg kg ) As

AsIII DMA MMA sugar 1 sugar 2 sugar 4 AB extraction efficiency (%) column recovery (%)

894 ( 157

98

2

51

97

340 ( 18

68

30

0.2

0.2

0.1

1

40

100

zooplankton(ø180) Long Lake phytoplankon

39 ( 7 396 ( 65

67 94

29 6

0.6

0.3

0.1

3

60 38

101 97

zooplankton (ø 45)

265 ( 60

87

12

0.5

36

91

zooplankton(ø180)

45 ( 2

64

33

0.6

0.1

0.3

2

57

95

Kam Lake phytoplankon zooplankton(ø45)

154 ( 1

92

8

50

103

zooplankton(ø180)

7(2

16

22

1

2

3

4

47

5

95

77

zooplankton(ø250)

11 ( 2

30

28

1

2

3

3

24

9

76

71

Grace Lake phytoplankon

Zooplankton samples were divided into different groups of size: organisms retained on 250 μm sieve (the largest ones), organisms retained on 180 μm sieve, and organisms retained on 45 μm sieve (the smallest ones).

a

show that Grace Lake can be used as a background, with low arsenic concentration in water and sediments. Long Lake, however, contains higher concentrations of arsenic, which is probably due to the naturally elevated arsenic concentration in the Yellowknife area. The mine-contaminated Kam Lake sites show a high arsenic content in water and sediments, although the contamination seems localized; position 3 (Figure 1), which was far from the contamination source, contains arsenic concentrations similar to those in Long Lake. Arsenic in Plankton. The zooplankton species collected in Long Lake and Kam Lake were mostly copepods from the genus Cyclops, at more than 90% of total zooplankton collected. The rest was a mix of Daphnia sp. and unidentified rotifers. Total arsenic in plankton was not significantly different at the three sampling locations of Kam Lake, so organisms from these three locations were combined into one sample labeled as Kam Lake plankton for speciation analysis. The uncontaminated Grace Lake contained Cyclops sp. at more than 95%. In Grace Lake, the organisms were more abundant and much larger than those from Kam Lake and Long Lake, with an average organism size of 500 μm, as compared to less than 250 μm in Kam and Long Lakes. Zooplankton organisms collected from the three lakes were sieved into different groups of size labeled ø250, ø180, and ø45. Arsenic concentrations in plankton are presented in Table 2. The results show that arsenic in phytoplankton ranges from 154 mg kg
1 in the background Grace Lake to 894 mg kg
1 in the highly contaminated Kam Lake. Zooplankton contains 7 mg kg
1 in Grace Lake, and ranges up to 340 mg kg
1 arsenic in small organisms (ø45) from Kam Lake. The standard deviation measured on triplicates is within 20% of the mean in most of the samples analyzed. These measurements were carried out on freeze-dried samples, and thus all concentrations are dry weight. The wet to dry weight ratio was measured as 30 for these zooplankton organisms, so that wet weight concentrations ranged from 0.2 to 11 mg kg
1 in zooplankton. These wet weight values are significantly higher than the arsenic concentrations measured in the water samples, indicating that plankton organisms accumulate arsenic from their environment. Table 2 indicates that small organisms accumulate more arsenic than larger organisms. Specifically, phytoplankton and small zooplankton organisms (