Spatial Distribution and Temporal Variability of Arsenic in Irrigated

Jul 18, 2007 - Swiss Federal Institute of Aquatic Science and Technology (Eawag), Ueberlandstrasse 133, CH-8600 Dübendorf, Switzerland, Institute of ...
0 downloads 9 Views 287KB Size
Environ. Sci. Technol. 2007, 41, 5960-5966

Spatial Distribution and Temporal Variability of Arsenic in Irrigated Rice Fields in Bangladesh. 1. Irrigation Water L I N D A C . R O B E R T S , †,‡ S T E P H A N J . H U G , * ,† J E S S I C A D I T T M A R , ‡ A N D R E A S V O E G E L I N , ‡ G A N E S H C . S A H A , §,| M. ASHRAF ALI,§ A. BORHAN M. BADRUZZAMAN,§ AND RUBEN KRETZSCHMAR‡ Swiss Federal Institute of Aquatic Science and Technology (Eawag), Ueberlandstrasse 133, CH-8600 Du ¨ bendorf, Switzerland, Institute of Biogeochemistry and Pollutant Dynamics, Department of Environmental Sciences, ETH Zurich, CHN, CH-8092 Zurich, Switzerland, and Department of Civil Engineering, Bangladesh University of Engineering and Technology, Dhaka-1000, Bangladesh

Around 38% of the area of Bangladesh is irrigated with groundwater to grow dry season crops, most importantly boro rice. Due to high As concentrations in many groundwaters, over 1000 tons of As are thus transferred to arable soils each year, creating a potential risk for future food production. We studied the reactions and changing speciation of As, Fe, P, and other elements in initially anoxic water during and after irrigation and the resulting spatial distribution of As input to paddy soils near Sreenagar (Munshiganj), 30 km south of Dhaka, in January and April 2005 and February 2006. The irrigation water had a constant concentration of 397 ( 7 µg L-1 As (∼84% AsIII), 11 ( 0.1 mg L-1 Fe, and 2 ( 0.1 mg L-1 P. During the fast flow along the longest irrigation channel (152 m) As, Fe, and P speciation changed, but total concentrations did not decrease significantly, indicating that As input to fields was independent of the length of the irrigation channels. In contrast, during slow water flow across the fields, As, Fe, and P concentrations decreased strongly with increasing distance from the water inlet, due to formation and settling of As- and P-bearing Fe aggregates and by adsorption to soil minerals. Total As concentrations in field water were ∼3 times higher close to the inlet than in the opposite field corner shortly after irrigation, and decreased to below 35 µg L-1 over the next 72 h. The laterally heterogeneous transfer of As, Fe, and P from irrigation water to soil has important consequences for their distribution in irrigated fields and needs to be considered in sampling and in assessing the dynamics and mass balances of As fluxes among irrigation water, soil, and floodwater.

* Corresponding author e-mail: [email protected]; phone: +41-448235454; fax: +41-44-8235210. † Swiss Federal Institute of Aquatic Science and Technology (Eawag). ‡ ETH Zurich. § Bangladesh University of Engineering and Technology. | Present address: Department of Civil Engineering, Dhaka University of Engineering and Technology, Gazipur 1700, Bangladesh. 5960

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 17, 2007

Introduction Naturally occurring high concentrations of arsenic in the groundwater in Bangladesh and West Bengal have raised considerable concern (1, 2). While providing access to safe drinking water remains an urgent problem, further longterm risks to human health and the environment may prove problematic. Since the late 1980s groundwater is increasingly being used for the irrigation of dry season crops, especially boro rice (3). In 2004, 71% of the total area under irrigation (i.e., 38% of the area of Bangladesh) was irrigated by groundwater, 54% from shallow tube wells, often particularly rich in As (4). Boro rice, which provides greater yields than varieties grown during the wet season, accounted for 48% of total rice production in 2004 (4). According to recent estimates, 1360 metric tons of arsenic are transferred to paddy fields in Bangladesh through irrigation each year (3). Evidence of As accumulation in soils (5-8) and uptake by rice plants (9, 10) has been presented in various studies. However, the long-term fate of As in irrigated rice fields, which is of great importance for the future of food production in Bangladesh, is presently poorly understood. A recent study shows topsoil As concentrations to increase during irrigation and decrease during monsoon flooding (11), while data from Ali et al. (12) suggests soil As concentrations may be highest in the vicinity of an irrigation well. Better knowledge of the spatial and temporal variability of As concentrations in the irrigation water-soil-plant system is crucial for drawing mass balances of As fluxes. We hypothesized that As distribution with irrigation water and input to soil may be laterally heterogeneous for the following reasons. First, aeration of anoxic groundwater containing dissolved FeII leads to Fe oxidation and formation of hydrous ferric oxide (HFO) colloids. At pH values of 7-8, AsV and phosphate are known to have high sorption affinity to precipitating HFO colloid aggregates, while AsIII exhibits a much lower sorption affinity (13). Depending on arsenic speciation and the As/P ratio, varying fractions of colloidal arsenic may thus form in the irrigation water. Aggregation and settling of As-bearing colloids may lead to decreasing As input to soils at greater distances from an irrigation well. Second, adsorption of As to the soil surface may have a similar effect. The objective of the present study was therefore to investigate the processes determining the distribution of As with irrigation water and to assess the resulting spatial heterogeneity in channels and fields irrigated by a single well. The study was conducted in conjunction with a soil study at the same field site (14), to permit reciprocal and direct validation of results.

Materials and Methods Field Site. The field site (Figure 1) is located near Sreenagar in Munshiganj district, 30 km south of Dhaka and 5 km north of the river Ganges, and comprises 18 paddy fields (total area 3.16 ha), exclusively used for the production of boro rice since the construction of the local irrigation well in the early 1990s. The area is characterized by a subtropical hot monsoon climate with intense annual flooding between June/July and November and is hydrologically well described (15). The irrigation well extracts groundwater from a depth of 30-60 m of an aquifer previously studied by Swartz et al. (16) with a pumping rate of approximately 19 L s-1. Fields are irrigated 18-21 times between early January and late April. Local boro rice production requires ∼1 m season-1 of irrigation water (3, 17), as confirmed by recorded quantities applied to this field site (11) and to a paddy field nearby (15) in 2004 (0.96 and 0.9 m, respectively). Shallow irrigation 10.1021/es070298u CCC: $37.00

 2007 American Chemical Society Published on Web 07/18/2007

TABLE 1. Well Water Composition at the Field Site: Mean Values ( Standard Deviations for All Measurements Taken in January 2005, April 2005, and February 2006 (n ) Number of Samples) total (n ) 27)

FIGURE 1. Map of the field site showing the sampling locations in the irrigation channel and in fields I and R (the secondary sampling points in field I were not sampled at all times). The arrows mark the irrigation water inlets of these fields.

channels lead to the fields enclosed by low embankments. For irrigation, a narrow breach (inlet) is opened in the embankment through which the water enters the field. The inlet locations remain unvaried over the years. Fields are irrigated for 2-4 h at a time until covered by 3-10 cm of water. Over the next days the water level gradually decreases until the moist soil surface is exposed. Scattered depressions filled with water may still be present in parts of the field, while soil cracks form in other parts. Field are thus subjected to periodic flooding and drying. Sampling Design. Irrigation water was sampled during three field campaigns: shortly after the onset of irrigation in January 2005, toward the close of the irrigation season in April 2005, and during the subsequent irrigation season in February 2006. The longest channel stretch within the field site area (distance well to inlet field R, 152 m, Figure 1) was sampled in January and April 2005. Water samples from fields I (2778 m2) and R (3057 m2), situated at different distances from the irrigation well (20 and 152 m) were collected shortly after the fields had been irrigated (start of sampling 0-30 min after irrigation) and at subsequent times (24, 48, and 72 h) on sampling grids shown in Figure 1. Collection of samples (from the center of the water column) required 1.5-2 h for each field. For clarity, irrigation water is referred to as well water, channel water, and field water according to the location of sampling. Collection of Samples. At all points, filtered (0.2 µm, nylon) and unfiltered samples were collected into preacidified 4 mL polypropylene vials (80 µL of 2 M HCl). Since the reliability of preserving AsIII-AsV speciation remains controversial (18, 19), an additional 2 mL was filtered through a 0.2 µm filter combined with a modified As speciation cartridge (20) as detailed in the Supporting Information (SI). Additional samples for total and dissolved organic carbon (TOC/DOC), alkalinity, and chloride measurements were collected as described in the SI. Conductivity, pH, and dissolved O2 were measured in the field with a multiparameter sensor (WTW, Germany). The following notation is used for the As fractions: total As (As(tot)), dissolved As (As(diss)), and dissolved AsIII (AsIII(diss), the latter both operationally defined by 0.2 µm-filtering. Dissolved AsV (AsV(diss) ) As(diss) - AsIII(diss)) and colloidal As (As(coll) ) As(tot) - As(diss)) were calculated by differences.

As (µg L-1) AsIII (µg L-1) Fe (mg L-1) Mn (µg L-1) P (mg L-1) Si (mg L-1) Ca (mg L-1) Mg (mg L-1) Na (mg L-1) K (mg L-1) S (mg L-1) Cl (mg L-1) TOC (mg L-1) alkalinity (mM) pH O2 (mg L-1) conductivity (µS cm-1) T (°C)

397 ( 7 11.0 ( 0.1 682 ( 13 1.97 ( 0.06 20.1 ( 0.2 99.2 ( 0.8 26.7 ( 0.2 16.3 ( 0.3 5.1 ( 0.2 0.37 ( 0.11

dissolved (n ) 27) 397 ( 7 332 ( 27 10.9 ( 0.2 679 ( 10 1.92 ( 0.07 20.1 ( 0.3 98.9 ( 0.9 26.6 ( 0.2 16.3 ( 0.3 5.1 ( 0.2 0.28 ( 0.18 9.3 ( 1.1(n ) 3)

7.7 ( 0.2 (n ) 4) 8.3 ( 0.3 (n ) 3) 7.0 ( 0.1 (n ) 10) < 1 (n ) 4) 786 ( 11(n ) 7) 26.1 ( 0.4 (n ) 21)

Chemicals and Analyses. Arsenic was measured by hydride generation atomic fluorescence spectroscopy (HG-AFS), Fe, P, Mn, and other elements were measured by inductively coupled plasma optical emission spectroscopy (ICP-OES), and Cl-, TOC/DOC, and alkalinity were measured by instrumentation and methods described in the SI. Details regarding the analytical procedures are given as SI.

Results and Discussion Composition of Well Water. No significant changes in well water composition were observed among the three sampling dates as shown in Tables 1 and S1. Total and dissolved concentrations did not differ significantly. Since TOC concentrations were comparatively small, phosphate can be regarded as the predominant P species present (Porg/Corg ∼1/106 (21), hence Porg in well water ∼0.19 mg L-1). Chemical Changes in Channel Water. Measurements carried out in January and April 2005 along the longest channel stretch (152 m) yielded very similar results. Mean channel flow velocity was 0.4 m s-1, translating into 6.3 min of flow time. Irrigation water remained unsaturated with oxygen, increasing from 4 to 5 mg L-1 O2 in the well pool to a maximum of 6 mg L-1 at the furthest channel point, while pH increased by 0.1 to 0.2 pH units. Figure 2 shows the concentrations of Fe, P, and As in channel water as a function of distance from the well. Dissolved Fe concentrations decreased by 76% along the channel, indicating that most Fe was oxidized and formed HFO colloids. Dissolved P and As concentrations also decreased, which can be attributed to sorption to HFO. Dissolved P concentrations decreased much more strongly than dissolved As concentrations (by 96% and 21%, respectively) due to significantly larger initial P concentrations (64 µM P vs 5 µM As), and to the prevalence of AsIII over AsV in the well water, both factors enhancing the adsorption competitiveness of phosphate with respect to As (13). Assuming all adsorbed arsenic to be AsV, 30% of the AsIII initially present was oxidized within the channel, in agreement with previous studies showing FeII oxidation to lead to partial As oxidation (13, 22). The stronger decrease in dissolved As concentrations within the last ∼40 m of channel (Figure 2C) is consistent with the P/As and AsIII/AsV ratios becoming more favorable for As sorption. In contrast to dissolved Fe, total Fe concentrations decreased only slightly within the channel (12%), showing that aggregation and settling of HFO colloids was limited. VOL. 41, NO. 17, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5961

FIGURE 2. Total and dissolved (0.2 µm-filtered) Fe (mg L-1), P (mg L-1), and As (µg L-1) concentrations, as well as dissolved AsIII concentrations (µg L-1) in channel water, sampled in January and April 2005. This implies that sorption of P and As to HFO, though occurring during channel flow, was of limited significance in terms of P and As removal. Consistently, total P and As concentrations decreased by only 11% and 8%, respectively, within the channel, and field water As concentrations measured close to the inlets of fields I and R did not differ significantly (Figure S3). Total As input to any field within the irrigation area is thus virtually unaffected by the length of the channel stretch between well and field inlet. Consistently, mean soil As concentrations measured in 8 fields in January 2005 were independent of the field location with respect to the irrigation well (Part 2, (14)). P input to the paddy fields is largely colloidal, while As input is predominantly dissolved AsIII. Dissolved Mn concentrations decreased by 8.5% within the irrigation channel (Figure S1). Since MnII oxidation rates are very low at neutral pH (21) it is unlikely that Mn oxides 5962

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 17, 2007

FIGURE 3. Dissolved O2 (mg L-1), pH, and total Ca concentrations (mg L-1) in field water on field I, 0.5-3.5 h after irrigation events in January 2005, April 2005, and February 2006. Irrigation times: 9:15-11:30 am (Jan 2005), 7:00-10:00 am (Apr 2005), 7:10-9:40 am (Feb 2006, 1), 1:45-4:30 pm (Feb 2006, 2). (A) Dissolved O2 (mg L-1) during hours of sunlight in January 2005 (2:45 pm), April 2005 (12:15 am), and February 2006 (12:00 am) and during the dark (6 pm) in February 2006. (B) pH and (C) total Ca concentrations (mg L-1) in January 2005 (1:15 pm), April 2005 (11:30 am), and February 2006 (4:45 pm). formed during channel flow. The decrease in dissolved Mn concentrations may be due to sorption of MnII to HFO or possibly to other colloidal phases. Chemical Changes in Field Water. Figure 3 shows dissolved O2, pH, and total Ca concentrations in field water shortly after several irrigation events in Field I. Oxygen levels showed a strong seasonal and diurnal variation linked to the availability of light for algal photosynthesis (Figure 3A) (23). During the early stages of plant growth in January 2005 and February 2006, the light-exposed field surface gave rise to clearly visible algae blooms and to O2 levels of up to 17 mg L-1 during the daytime. In April 2005, when the plant canopy

FIGURE 5. Lateral As distribution (µg L-1) in irrigation water on field I shortly (0-2 h) after a 2.75-h irrigation period in February 2006. Concentrations of As(tot), As(diss), and AsIII(diss) are represented by the areas of the outer black, middle white, and inner gray circles, respectively. As(tot) concentrations are also shown in numbers.

FIGURE 4. Irrigation water on field I shortly (0-2 h) after a 2.75-h irrigation period in February 2006. Total and dissolved (0.2 µmfiltered) Fe (mg L-1) (A), P (mg L-1) (B), and As(tot) (µg L-1) concentrations, as well as dissolved AsIII (µg L-1) concentrations (C) are shown. was efficiently shading the soil surface, as well as after dusk, O2 levels ranged between 4 and 8 mg L-1. pH values increased with distance from the water inlet (Figure 3B), which is attributable to CO2 degassing from the originally CO2-supersaturated groundwater. The overall higher pH values measured in January and February as compared to April (Figure 3B) are likely to reflect the higher photosynthetic activity in the unshaded water (23). Additionally, in April degassing of CO2 from the field water may have been slower due to the wind shelter provided by the plant canopy. Ca concentrations decreased with increasing distance from the inlet (Figure 3C) and with time after irrigation. Since the irrigation water is oversaturated with respect to CaCO3, CO2 degassing and the concomitant rise in pH may cause formation of CaCO3 and/or adsorption of Ca on soil particles. Any formation of CaCO3 seems to have taken place by surface precipitation rather than within the water body, since there

were no significant differences between total and dissolved Ca concentrations (not shown). Mg showed a behavior similar to Ca, and conductivity also decreased with increasing distance from the water inlet and time after irrigation (not shown). TOC concentrations showed a slight increase with distance from the inlet but did not appear to increase over time (Figure S2). Since differences between DOC and TOC concentrations were small (not shown) desorption rather than resuspension of soil organic matter seems to have taken place during irrigation, in agreement with the slowness of field water flow and the low degree of turbulence observed. Lateral Distribution and Speciation of Fe, P, and As in Field Water Shortly after Irrigation. Figure 4 shows Fe, P, and As concentrations on field I after a 2.75-h irrigation period in February 2006. Samples taken on fields I and R following other irrigation events showed similar distributions (see Figure S3 for As(tot) distributions). As is apparent from Figure 4A, all FeII was oxidized and formed HFO colloids within a distance of 20 m from the field inlet. Dissolved P concentrations were likewise very low in all field points ( 50 µg L-1 in Bangladesh (199 ( 166 µg L-1 As, 5.3 ( 4.8 mg L-1 Fe, and 1.47 ( 1.48 mg L-1 P; 1). The Fe/As and

As/P molar ratios at our field site, more relevant in terms of resulting spatial heterogeneity, are higher than in 55% and 67%, respectively, of these groundwaters (1). High As/P ratios, resulting in less phosphate competing for sorption sites on ferric colloids and on soil components can be expected to lead to more heterogeneous As distributions. High Fe/As ratios, leading to larger fractions of colloidal As may also enhance lateral heterogeneity. The heterogeneity observed at the field site may thus be slightly more pronounced than on average. The irrigation system in place at the study site, involving channels leading to each field, is representative of Munshiganj district and of many other districts in Bangladesh (Comilla, Brahmanbaria, Faridpur, Bogra, Naogaon) (11, 30). Depending on the size of the area irrigated by a well, differences in total As input may become apparent between single fields, since longer irrigation channels would result in more HFO aggregates settling during channel flow. The size range of areas irrigated by single wells reported in other studies (1.755.25 ha (11); 1-5 ha (6)) indicates that most areas are probably not large enough for HFO settling to increase substantially in channels, however. If there are areas where irrigation is achieved by a cascade system, in which the water is directed through a succession of adjacent fields, the most severe As input would be expected in the field first receiving the irrigation water. In principle, passage of the irrigation water through a designated treatment field or pond prior to its distribution might offer a simple means for reducing As input to soils used for crop production. The considerations outlined above indicate that local irrigation practices need to be considered carefully to design meaningful water and soil sampling campaigns. Knowledge of the pathways of As input to paddy soils via irrigation also provides a basis for addressing the reverse processes of As remobilization from the soil, which other studies, including Part 2 of this work, have shown to take place during monsoon flooding (12, 14, 30). Understanding the vertical redistribution of As to deeper layers and/or into the floodwater, and the lateral export of As with floodwater to rivers and into the ocean, is of crucial importance for the assessment of longterm risks of As accumulation in soils in Bangladesh.

Acknowledgments We thank T. Ruettimann for analytical support, the technical staff from Bangladesh University of Engineering and Technology and Sojib Chowdhury from Bashailbhog village for help with field work, and the people of Bashailbhog village for access to their paddy fields. Funding of this research by the Swiss National Science Foundation (Grant 200021105612/1 and 200020-113654/1) is gratefully acknowledged.

Supporting Information Available Details regarding analytical procedures and well water composition, Mn and TOC concentrations, As concentrations in field R, lateral As distribution after irrigation of field I from a temporarily shifted inlet, and AsV(diss) concentrations in field I over time. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) British Geological Survey (BGS) and Bangladesh Department of Public Health Engineering (DPHE)Arsenic contamination of groundwater in Bangladesh, Volume 2: Final report. Kinniburgh, D. G.; Smedley, P. L., Eds. BGS Report WC/00/19 BGS: Keyworth, UK., 2001. (2) Smith, A. H.; Lingas, E. O.; Rahman, M. Contamination of drinking water by arsenic in Bangladesh: A public health emergency. Bull. WHO 2000, 78, 1093-1103.

(3) Ali, M. A.; Badruzzaman, A. B. M.; Jalil, M. A.; Hossain, M. D.; Ahmed, M. F.; Masud, A. A.; Kamruzzaman, M.; Rahman, M. A. Fate of arsenic extracted with groundwater. In Fate of Arsenic in the Environment; Ahmed, M. F., Ed.; ITN International Training Network: Dhaka, 2003; pp 7-20. (4) Handbook of Agricultural Statistics; Ministry of Agriculture: Bangladesh, 2005. (http://www.moa.gov.bd/statistics/statistics. htm). (5) Alam, M. B.; Sattar, M. A. Assessment of arsenic contamination in soils and waters in some areas of Bangladesh. Water Sci. Technol. 2000, 42, 185-193. (6) Meharg, A. A.; Rahman, M. M. Arsenic contamination of Bangladesh paddy field soils: Implications for rice contribution to arsenic consumption. Environ. Sci. Technol. 2003, 37, 229234. (7) Norra, S.; Berner, Z. A.; Agarwala, P.; Wagner, F.; Chandrasekharam, D.; Stuben, D. Impact of irrigation with As rich groundwater on soil and crops: A geochemical case study in West Bengal Delta Plain, India. Appl. Geochem. 2005, 20, 18901906. (8) van Geen, A.; Zheng, Y.; Cheng, Z.; He, Y.; Dhar, R. K.; Garnier, J. M.; Rose, J.; Seddique, A.; Hoque, M. A.; Ahmed, K. M. Impact of irrigating rice paddies with groundwater containing arsenic in Bangladesh. Sci. Total Environ. 2006, 367, 769-777. (9) Abedin, M. J.; Cotter-Howells, J.; Meharg, A. A. Arsenic uptake and accumulation in rice (Oryza sativa L.) irrigated with contaminated water. Plant Soil 2002, 240, 311-319. (10) Abedin, M. J.; Cresser, M. S.; Meharg, A. A.; Feldmann, J.; CotterHowells, J. Arsenic accumulation and metabolism in rice (Oryza sativa L.). Environ. Sci. Technol. 2002, 36, 962-968. (11) Saha, G. C.; Ali, M. A. Dynamics of arsenic in agricultural soils irrigated with arsenic contaminated groundwater in Bangladesh. Sci. Total Environ. 2006, 379, 180-189. (12) Ali, M. A.; Badruzzaman, A. B. M.; Jalil, M. A.; Hossain, M. D.; Ahmed, M. F.; Masud, A. A.; Kamruzzaman, M.; Rahman, M. A. Arsenic in plant-soil environment in Bangladesh. In Fate of Arsenic in the Environment; Ahmed, M. F., Ed.; ITN International Training Network: Dhaka, 2003; pp 85-112. (13) Roberts, L. C.; Hug, S. J.; Ruettimann, T.; Billah, M. M.; Khan, A. W.; Rahman, M. T. Arsenic removal with iron(II) and iron(III) in waters with high silicate and phosphate concentrations. Environ. Sci. Technol. 2004, 38, 307-315. (14) Dittmar, J.; Voegelin, A.; Roberts, L. C.; Hug, S. J.; Saha, G. C.; Ali, M. A.; Badruzzaman, A. B. M.; Kretzschmar, R. Spatial distribution and temporal variability of arsenic in irrigated rice fields in Bangladesh. 2. Paddy soil. Environ. Sci. Technol. 2007, 41, 5967-5972. (15) Harvey, C. F.; Ashfaque, K. N.; Yu, W.; Badruzzaman, A. B. M.; Ali, M. A.; Oates, P. M.; Michael, H. A.; Neumann, R. B.; Beckie, R.; Islam, S.; Ahmed, M. F. Groundwater dynamics and arsenic contamination in Bangladesh. Chem. Geol. 2006, 228, 112136. (16) Swartz, C. H.; Blute, N. K.; Badruzzman, B.; Ali, A.; Brabander, D.; Jay, J.; Besancon, J.; Islam, S.; Hemond, H. F.; Harvey, C. F. Mobility of arsenic in a Bangladesh aquifer: Inferences from geochemical profiles, leaching data, and mineralogical characterization. Geochim. Cosmochim. Acta 2004, 68, 4539-4557. (17) Field, D. I. Land Evaluation for Crop Diversification; Crop Diversification Programme (CDP): Dhaka, Bangladesh, 1995. (18) Bednar, A. J.; Garbarino, J. R.; Ranville, J. F.; Wildeman, T. R. Preserving the distribution of inorganic arsenic species in groundwater and acid mine drainage samples. Environ. Sci. Technol. 2002, 36, 2213-2218. (19) McCleskey, R. B.; Nordstrom, D. K.; Maest, A. S. Preservation of water samples for arsenic(III/V) determinations: An evaluation of the literature and new analytical results. Appl. Geochem. 2004, 19, 995-1009. (20) Meng, X.; Wang, W. Speciation of arsenic by disposable cartridges. In Third International Conference on Arsenic Exposure and Health Effects, San Diego, California, 1998. (21) Stumm, W.; Morgan, J. Aquatic Chemistry. Chemical Equilibria and Rates in Natural Waters, 3rd ed.; John Wiley & Sons: New York, 1996. (22) Hug, S. J.; Leupin, O. Iron-catalyzed oxidation of arsenic(III) by oxygen and by hydrogen peroxide: pH-dependent formation of oxidants in the Fenton reaction. Environ. Sci. Technol. 2003, 37, 2734-2742. (23) Kirk, G. The Biogeochemistry of Submerged Soils; John Wiley & Sons: Chichester, U.K., 2004. (24) Wilkie, J. A.; Hering, J. G. Rapid oxidation of geothermal arsenic(III) in streamwaters of the Eastern Sierra Nevada. Environ. Sci. Technol. 1998, 32, 657-662. VOL. 41, NO. 17, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5965

(25) Scott, M. J.; Morgan, J. J. Reactions at oxide surfaces. 1. Oxidation of as(III) by synthetic birnessite. Environ. Sci. Technol. 1995, 29, 1898-1905. (26) Voegelin, A.; Hug, S. J. Catalyzed oxidation of arsenic(III) by hydrogen peroxide on the surface of ferrihydrite: An in situ ATR-FTIR study. Environ. Sci. Technol. 2003, 37, 972-978. (27) Buschmann, J.; Canonica, S.; Lindauer, U.; Hug, S. J.; Sigg, L. Photoirradiation of dissolved humic acid induces arsenic(III) oxidation. Environ. Sci. Technol. 2005, 39, 9541-9546. (28) Hug, S. J.; Canonica, L.; Wegelin, M.; Gechter, D.; von-Gunten, U. Solar oxidation and removal of arsenic at circumneutral pH in iron containing waters. Environ. Sci. Technol. 2001, 35, 21142121.

5966

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 17, 2007

(29) Goldberg, S. Competitive adsorption of arsenate and arsenite on oxides and clay minerals. Soil Sci. Soc. Am. J. 2002, 66, 413421. (30) Saha, G. C. Accumulation of arsenic in agricultural soil and selected crops. Ph.D. thesis, Bangladesh University of Engineering and Technology, 2006.

Received for review February 6, 2007. Revised manuscript received June 4, 2007. Accepted June 12, 2007. ES070298U