Environ Geochem Health DOI 10.1007/s10653-015-9710-3
ORIGINAL PAPER
Assessing and simulating the major pathway and hydrogeochemical transport of arsenic in the Beitou– Guandu area, Taiwan Chen-Wing Liu . Chin-Jen Wang . Yu-Hsiun Kao
Received: 24 October 2014 / Accepted: 7 May 2015 Ó Springer Science+Business Media Dordrecht 2015
Abstract This study involved assessing and simulating the probable major pathways (surface and subsurface flow) and hydrogeochemical transport of arsenic (As) in the Beitou–Guandu area, Taiwan. A onedimensional (1-D) generic, reactive, chemical transport model (PHREEQC) was adopted. The calibrated model showed that As transported to the downstream Guandu plain and Tan Shui river mouth accounted for 50.7 and approximately 100 % of the As in the subsurface flow pathway, respectively, suggesting that subsurface flow constituted a major As pathway. The highest As water concentration occurred near the Beitou geothermal valley because of the low pH and high redox potential in both the surface and subsurface pathways. However, As may be scavenged by aqueous Fe(II) in a reducing environment. The As concentrations in the downstream Guandu plain and Guandu wetland decreased as the simulated time increased, resulting in the adsorption of As on the surface of Fe oxydroxides and limiting the mobility of As in the surface flow pathway. The major retardation mechanism of As mobility in the subsurface flow pathway of the Guandu plain and Guandu wetland was governed by the adsorption reactions of iron-oxide and ironsulfide minerals. The 1-D transport model was applied to predict the evolution of As in the subsurface flow C.-W. Liu (&) C.-J. Wang Y.-H. Kao Department of Bioenvironmental Systems Engineering, National Taiwan University, Taipei 10617, Taiwan e-mail:
[email protected]
pathway from 2013 to 2020. The results indicated that the As concentrations in all cells gradually increased. The geochemical redox reactions of As in the subsurface pathway subsequently led to the oxidization of As-bearing sulfides, causing As concentrations to rise substantially in the hillside area. Keywords Arsenic (As) Geothermal PHREEQC Pathway pe-pH diagram
Introduction Geothermal water often contains relatively high contents of arsenic as a result of the leaking and dissolving arsenic from rock minerals, which occurs mostly in the geothermal reservoir at high temperature (Smedley and Kinniburgh 2002; Yoshizuka et al. 2010). The discharge of arsenic (As)-rich geothermal spring waters through volcanic activity may result in As pollution of groundwater, surface water, downstream wetlands, and estuary ecosystems (Lie`vremont et al., 2009). This discharge has caused severe environmental problems in several geothermal areas, including those in Taiwan, Japan, New Zealand, Iceland, Chile, the Kamchatka Peninsula, and parts of the United States (White et al. 1963; Welch et al. 1988; Ball et al. 1998; Yoshizuka et al. 2010; Sigfusson et al. 2011). Kelepertsis et al. (2001)reported that arsenic pollution of soil in the Susaki area of Greece was due to the volcanic activity and
123
Environ Geochem Health
Fig. 1 a Study area and b the schematic diagram of 1-D model for As transport pathways
geothermal water. Alexakis and Gamvroula (2014) also found that high arsenic contents in the OroposKalamos lignite result in high arsenic concentrations in the groundwater. Yoshizuka et al. (2010) surveyed 16 sites of hot spring areas in Kyushu, Japan, and the arsenic concentration is ranged from 110 to 3230 lg/L. Active As-rich geothermal systems preferentially concentrate As in the aqueous phase primarily as
123
As(OH)3. Moreover, both of the thioarsenite and thioarsenate species in natural geothermal solutions depend on the temperature, pH, redox potential, and H2S content (Pokrovski et al. 1996; Tsai et al. 2009). The Beitou–Guandu area includes the Beitou geothermal valley, Guandu plain, and Guandu wetland in Northern Taiwan and covers an area of approximately 100 km2 (Fig. 1a). The Beitou
Environ Geochem Health
geothermal valley is located in the Tatun volcanic area, where large-scale andesitic volcanic eruptions occurred during the Pleistocene epoch. Huang Gang creek and Kuei Tzu Keng creek flow through the valley. The As concentration of hot spring water in the Beitou geothermal valley was up to 4.6 mg/L which markedly exceeded the limits specified in the drinking water guidelines of the World Health Organization (WHO; 0.01 mg/L; Wang et al. 2011). Acidic geothermal spring water with high As and sulfate concentrations is discharged into the downstream alluvial plain and wetland (Chiang et al. 2010; Kao et al. 2013). Guandu plain is situated at the intersection of the Tan Shui river and Kee Lung river, northwest of the Taipei basin. The geomorphic features, namely the flat topography and low altitude (0–6 m), are a typical floodplain (approximately 1300 ha). The Guandu plain is an area where rice is cultivated and contains the only large-scale agricultural rice fields remaining in metropolitan Taipei. Long-term irrigation with water high in As has caused substantial accumulation of As in the surface (0–15 cm) and subsurface (15–30 cm) soil in the Guandu plain (up to 145 and 143 mg/kg, respectively). The soil contamination standard of the Taiwan EPA is 60 mg/kg (Chiang et al. 2010). Moreover, the As content of downstream groundwater was as high as 42–49 and 11–25 lg/L in shallow and deep monitoring wells with depths of 20–30 and 50–60 m, respectively. The Guandu wetland is located downstream of the Guandu plain, along the Kee Lung river, and only 10 km from the Tan Shui river estuary, and the area is widely affected by tidal fluctuations. Tidal seawater can intrude into the upper estuary approximately 25 km from the river mouth and mix with the river water (Liu et al. 2001). Arsenic may gradually move downstream to the Guandu wetland and estuarine systems through surface water flow and groundwater recharge. According to the preliminary investigations by the Environmental Protection Bureau of Taipei County, the average As contents of mullet and catfish in the Tan Shui river estuary, located downstream of the Guandu wetland were 0.76 and 3.7 mg/kg (wet weight), respectively. Hence, high-As-content water in geothermal spring areas moves downstream to coastal plains, wetlands, and estuary ecosystems, causing concern for the safe use of spring water and potential hazardous effects on the environment. Most previous studies have examined the As content in paddy soil and crops in the
Guandu plain by using chromatographic methods (Syu et al. 2013; Chiang et al. 2013). The distribution and movement of As through various transport pathways are not adequately understood. This study assessed and simulated the probable pathways of the hydrogeochemical transport of As in the Beitou–Guandu area, Taiwan. The one-dimensional (1-D) transport model PHREEQC (Parkhurst and Appelo 1999) and graphical program Phreeplot (Kinniburgh and Cooper 2004) were adopted to simulate the transport of As species and the reactive flow pathway in the predominance stability field. Geochemical modeling involved quantifying the As speciation and identifying the formation of the corresponding mineral assemblage phase along the reactive flow path. The results are valuable in managing the environmental risk in areas exhibiting excessive amounts of As.
Materials and methods Conceptual model development Previous studies have indicated that the As concentration in hot spring water in the Beitou geothermal valley reached 4.6 mg/L (Wang et al. 2011). The downstream Guandu plain, Guandu wetland, and estuary ecosystems may be influenced by the toxicological effects of As. The transport pathways of As from the geothermal valley to the Tan Shui river mouth may be surface and subsurface pathways. The software PHREEQC is widely used for simulating geochemical reactions and transport processes in natural and polluted water. The program is based on the equilibrium chemistry of aqueous solutions interacting with minerals, gases, solid solutions, exchangers, and sorption surfaces. PHREEQC can model various 1-D transport processes including diffusion, advection, and dispersion. These processes can be incorporated with equilibrium and chemical kinetic reactions (Parkhurst and Appelo 1999). In this study, 1-D PHREEQC Interactive 2.17, based on the WATEQ4F thermodynamic database, was used to assess and simulate the behavior and hydrogeochemical transport of As in surface and subsurface flow pathways. A conceptual model was designed to assess the apportionment of As transported from the upstream Beitou area to the downstream Tan Shui river mouth through surface and subsurface flow pathways.
123
Environ Geochem Health Table 1 Initial water quality of geothermal spring, surface water and subsurface water Source/pathway
Zone
pH
Eh (mV)
As (mg/L)
Fe (mg/L)
SO42- (mg/L)
Spring
–
3
0
4.6
48.91
2429
Surface flow
Beitou geothermal valley
3.5
236
3
6.4
3534
Hillside Guandu plain
5.33 7
118 -5.9
0.03 0.016
1.42 3.54
108.7 25.51
Guandu wetland
7.01
-59
0.00268
0.89
786.25
Tan Shui river mouth
7
236
0.003
–
–
Infiltration
3
0
4.6
48.91
2429
Beitou geothermal valley
5.75
-64
0.544
3.38
85.92
Hillside
6.85
-64
0.209
3.38
85.92
Guandu plain
7.44
-163
0.054
0.61
9.9
Guandu wetland
7.6
-202
0.0105
6.69
867
Tan Shui river mouth
7
236
0.003
–
–
Subsurface flow
Table 2 As adsorption reactions and their thermodynamic constants
Adsorption reactions H3AsO4 ? Hfo_sOH = Hfo_sH2AsO4 ? H2O
8.61
H3AsO4 ? Hfo_wOH = Hfo_wH2AsO4 ? H2O
8.61
H3AsO4 ? Hfo_sOH = Hfo_sHAsO4- ? H? ? H2O
2.81
H3AsO4 ? Hfo_wOH = Hfo_wHAsO4- ? H? ? H2O
2.81
H3AsO4 ? Hfo_sOH = Hfo_sOHAsO43- ? 3H?
-10.12
H3AsO4 ? Hfo_wOH = Hfo_wOHAsO43- ? 3H?
-10.12
H3AsO3 ? Hfo_sOH = Hfo_sH2AsO3 ? H2O
5.41
H3AsO3 ? Hfo_wOH = Hfo_wH2AsO3 ? H2O
5.41
Figure 1b shows a schematic diagram of the 1-D As transport model. The input water quality of each area is shown in Table 1. For the surface flow pathway, 59 cells were discretized. Cells 1–3 represented the Beitou geothermal spring and were 600 m in length; Cells 4–8 represented the hillside and were 1000 m in length; Cells 9–43 represented the Guandu plain and were 2800 m in length; Cells 44–58 represented the Guandu wetland and were 1200 m in length; and Cell 59 represented the Tan Shui river mouth and was 10 m, where the groundwater is mixed with the Tan Shui river water. The spring water in Cell 1 continuously flowed to the lowermost Cell 59 and equilibrated with the minerals through surface sorption. The surface properties of the sediment were a specific surface area of 600 m2/g and site densities of 0.2 mol for Hfo_wOH and 0.005 mol for Hfo_sOH (Dzombak and Morel 1990). The As adsorption reactions are listed in Table 2. The time step was set to 0.1 day and
123
Log K
varied according to the surface flow velocity. The average surface flow velocity was 950 m/d (Yang 2011). The surface system length is set to 5610 m. Thus one pore volume in the system was 5.9 days [5610 (m)/950 (m/d)]. Dispersivity and diffusion were ignored because of the high surface flow velocity. For the subsurface flow pathway, 118 cells were discretized. Cell 1 modeled the geothermal spring water infiltration process and was 10 m in length; Cells 2–7 modeled the Beitou geothermal valley and were 600 m in length; Cells 8–17 modeled the hillside and were 1000 m in length; Cells 18–87 modeled the Guandu plain and were 2800 m in length; Cells 88–117 modeled the Guandu wetland and were 1200 m in length; Cell 118 modeled the Tan Shui river mouth and was 10 m in length. The physical properties of the surface (specific surface area: 15 m2/g; site density: 3.7 sites/nm2) were derived from those described by Bostick and Fendorf (2003). Surface sorption reactions of As were the same as those used in
Environ Geochem Health
the surface flow simulation (Table 2). The time step in each cell was specified as 20 days according to the hydraulic conductivity (approximately 2 m/d). The average groundwater flow velocity was 2.38 m/d (Yang, 2011). The subsurface system length is set to 5620 m. Consequently, one pore volume in this system was 2,360 days (approximately 6.5 years). The dispersivities in infiltration, the Beitou geothermal valley, the hillside, Guandu plain, Guandu wetland, and the Tan Shui river mouth were specified as 0.1, 1, 1, 0.8, 0.8, and, 0.1 m, respectively (Gelhar et al. 1992). The diffusion coefficient was based on the default value of PHREEQC, 3 9 10-10 m2/s.
Results Model calibration In the study area, water quality data recorded by the Taiwan EPA in the Tan Shui river mouth were used for model calibration. According to Fig. 2, the total As concentrations in the Tan Shui river mouth in the surface flow and subsurface flow pathway were 0.0042 and 2.11 lg/L, respectively. The sum of the As concentrations in the surface pathway and subsurface pathway was 2.11 lg/L, which is close to the 3 lg/L observed at the Tan Shui river mouth in 2013. The 1-D transport model was thus calibrated. Assessing the main transport pathway of As
Fig. 2 The simulated spatial–temporal profiles of As in 2007 and 2013 a surface flow pathway and b subsurface flow pathway
The apportionment of As transport through the surface and subsurface flow pathways in 2007 and 2013 was assessed. In 2007, the As in the Beitou geothermal valley, hillside, Guandu plain, Guandu wetland, and Tan Shui river mouth accounted for 74.6, 98.3, 0.1, 0.0031, and 0.0028 % of the As transported through the surface flow pathway, respectively. The As in the Beitou geothermal valley, hillside, Guandu plain, Guandu wetland, and Tan Shui river mouth accounted for 25.4, 1.72, 99.9, approximately 100, and approximately 100 % of the As transported through the subsurface flow pathway, respectively. The simulation results indicated that the surface flow pathway was the main As transport pathway only in the upstream regions; the subsurface flow pathway became the major As transport pathway in the mid-to-downstream areas. Moreover, in 2013, the As in the Beitou
geothermal valley, hillside, Guandu plain, Guandu wetland, and Tan Shui river mouth accounted for 49.6, 81.3, 49.3, 0.0003, and 0.002 % of the As transported through the surface flow pathway, respectively. Conversely, the As in the Beitou geothermal valley, hillside, Guandu plain, Guandu wetland, and Tan Shui river mouth accounted for 50.4, 18.7, 50.7, approximately 100, and approximately 100 % of the As transport through the subsurface flow pathway. As the simulation time increased, the apportionment of As transport through the subsurface flow pathway gradually increased, except in the Guandu plain. The high surface flow velocity caused the As species in the surface flow pathway to move faster than those in the subsurface flow pathway; therefore, the As apportionment in the subsurface flow pathway decreased in the Guandu plain. The results showed that subsurface flow
123
Environ Geochem Health Fig. 3 The pe-pH diagrams containing a the dominant As species and b the dominant Fe species for all the cells of surface flow pathway in February 2007
pathways was the main path of As transport in the Beitou–Guandu plain and Guandu wetland (including the Tan Shui river mouth). Fate and transport of As in the surface flow pathway Figure 2a shows the distribution of the As concentration throughout the surface flow pathway determined using the PHREEQC 1-D transport model. The
123
simulated As concentrations reached 4750 lg/L in Cells 1–8 because of the high As source recharge from the Beitou geothermal spring in 2007. To clarify the complex geochemical conditions during the As transport process in each area potentially formed through dissolution or precipitation or sorption by the solid mineral surface, predominance diagrams for the As–Fe–S system under standard conditions (25 °C and 1 atm) and at the initial water quality concentrations listed in Table 1 were constructed (Fig. 3a, b).
Environ Geochem Health Fig. 4 The pe-pH diagrams containing a the dominant As species and b the dominant Fe species for all the cells of surface flow pathway in July 2013
The simulated surface water had a low pH and a high As concentration in the aqueous Fe(II) stability area (Fig. 3b). Although previous studies have reported low mobility of As(III) under an acidic condition, because of the difference in electrical surface charge between aqueous As and solid minerals (Manning and Goldberg 1996; Sigfusson et al. 2008), the low pH corresponding to high ion solubility observed in this study may result in the dissolution of iron minerals; the
mobility of As is thus only slightly affected. However, the As concentration substantially decreased to 10 lg/ L in Cell 9 (end of the hillside area; Fig. 2a); the pH of the surface flow pathway gradually became near neutral, and the pathway became an oxidative environment, causing As(V) and Fe(III) to be dominant species (Fig. 3). The Fe(III) in an oxidative condition may precipitate to Fe oxydroxides, and its zero-point charge (pHzpc) was in the range of 6–8. The adsorption
123
Environ Geochem Health Fig. 5 The pe-pH diagrams containing a the dominant As species and b the dominant Fe species for all the cells of subsurface flow pathway in February 2007
of As to Fe oxydroxides is thus favorable because pH \ pHzpc (Maji et al. 2011). A large amount of As may be sorbed onto the surface of Fe oxydroxides (Fig. 3a). In the Guandu plain (Cells 9–43) and Guandu wetland (Cells 44–58), the As concentrations were within 10–0.21 lg/L and 0.19–0.028 lg/L, respectively. The redox potential values decreased as the distance from the Beitou geothermal valley increased. Moreover, the As concentrations of all
123
cells in these areas were positioned near the boundaries among the Hfo_wH2AsO3, H3AsO3, and AsS(OH)(HS)- stability fields (Fig. 3a) and near the boundary between Fe oxydroxides and iron sulfides (Fig. 3b); the transport of As was limited by the sorption reactions of the above minerals. Finally, at the end of the transport pathway, the Tan Shui river mouth, the simulated As concentration dropped to 0.029 lg/L.
Environ Geochem Health
Fig. 6 The pe-pH diagrams containing a the dominant As species and b the dominant Fe species for all the cells of subsurface flow pathway in July 2013
In 2013, the simulated As concentrations in the Beitou geothermal valley, hillside, Guaudu plain, Gandu wetland, and Tan Shui river mouth were 4560; 4550–4570; 0.0034–4560; 0.0045–0.0042; and 0.0042 lg/L, respectively. As the simulation time increased, the spring water from the Beitou geothermal valley continuously discharged into the
downstream cell, causing the system to have a low pH and a high redox potential. In Fig. 4a, the dominant species of As in Cells 1–9 (Beitou geothermal valley and hillside area) and Cells 10–118 (downstream plain, wetland and estuary) were aqueous As(III) and sorbed As(V), respectively. The simulated surface waters with low pH and redox potential were in the
123
Environ Geochem Health
Fig. 7 The Simulated As concentration profile by ?10 %, unchange, -10 % of the initial water quality concentration in the subsurface flow pathway in February 2020
Fe2? stability field and exhibited high As concentrations. By contrast, the simulated surface waters in the Fe oxydroxide field exhibited low As concentrations because of As adsorption on the surface of the minerals. Furthermore, the sharp decrease in the As concentration in 2013 was retarded in contrast to the spatial distribution of As in 2007 (Fig. 2a). Moreover, the As concentration of the downstream Guandu plain, Guandu wetland, and Tan Shui river mouth decreased as the simulation time increased because of the water dilution effect. Fate and transport of As in the subsurface flow pathway Figure 2b shows the profile of As concentrations in the subsurface flow pathway. The dominant species of As and Fe were As(III) and Fe(II), respectively, because of the reducing environment. In 2007, the simulated As concentrations in filtration, the Beitou geothermal valley, hillside, Guandu plain, and Guandu wetland were 5400; 252–5400; 72–120; 54–81; and 9–52 lg/L, respectively. The simulated As concentration at the end of the transport pathway (Tan Shui river mouth) was 10.54 lg/L, which exceeded the permissible limits for drinking water (10 lg/L) recommended by the WHO. A sharp decrease in the As concentration occurred in Cell 7 (end of the Beitou geothermal valley). Before Cell 7, the simulated subsurface water exhibited a high As content because aqueous Fe(II)
123
species were predominant (Fig. 6). After Cell 7, the composition of simulated subsurface water was near the boundary between the Hfo_wH2AsO3 H3AsO3 and AsS(OH)(HS)- stability fields and the boundary between Fe oxydroxides and iron sulfides (Fig. 5); the transport of aqueous As species was limited by the sorption reactions of the aforementioned minerals and the formation of thioarsenic. In 2013, the simulated As concentrations in infiltration, the Beitou geothermal valley, hillside, Guaudu plain, Gandu wetland, and Tan Shui river mouth were 4600; 4,600–4630; 167–4490; 65–236; 2.4–52; and 2.12 lg/L, respectively. Notably, the simulated As concentration in the Beitou geothermal valley and hillside are close to the As concentration of hot spring area in Kyushu, Japan(Yoshizuka et al. 2010). Comparing the aqueous As profiles for 2007 with those for 2013 revealed two dissimilarities: (1) the sharp decrease in the As concentration became more gradual in Cell 10 (hillside area; Fig. 2b); and (2) all redox potentials of the Guandu plain in 2007 and 2013 were under Fe reducing and sulfate reducing conditions (Stumm and Morgan 1996). In 2013, the As concentration increased to 124 lg/L because of the reductive dissolution of As-containing Fe oxydroxides. However, the As concentration decreased to 4.89 lg/L under the sulfate reducing condition in the Guandu wetland, because As(III) species were readsorbed by goethite and were constrained by the precipitation of sulfide minerals. We have performed a sensitivity analysis by varying the initial water quality concentrations ±10 %. The simulated results show only small changes of As concentrations of all three cases. The simulated As concentrations of ?10 %, unchanged, -10 % cases in the Tan Sui river mouth were 2.13, 2.12, 2.11 lg/L. Prediction of As transport in the Beitou–Guandu area According to the surface complexation reactions (Table 2), the 1-D PHREEQE transport model was applied to simulate As mobilization through the subsurface flow pathway (which was considered the main As transport pathway in this system). The simulation period was July 2013 to February 2020. In 2020, the As concentrations in all cells markedly increased (compare Figs. 7, 2b). The simulated total
Environ Geochem Health Fig. 8 The forecasting pepH diagrams containing a the dominant As species and b the dominant Fe species for all the cells of subsurface flow pathway in February 2020
As concentration in the Tan Shui river mouth through the subsurface flow pathway was 5.7 lg/L. The highest As concentration (2.94 mM) occurred in the hillside area and was attributed to the oxidation of Asbearing sulfides, which caused the As concentrations to increase. It is of interest to assess the predicted 2020 As concentration distribution by changing the initial water quality parameters ±10 %. Figure 7 show the
simulated As concentration distribution of all three cases in 2020. By increasing the initial water quality concentrations 10 %, the peak As concentration occurs in the edge of the hillside area and is moving faster than the parameters unchanged case. However, by decreasing the initial water quality concentrations 10 %, no wave-shape peak As concentration develops. Because the simulated redox potential of the hillside area is switched from oxidation to reducing condition
123
Environ Geochem Health
in the decreasing 10 % case which prohibits the oxidation of As-bearing sulfide dissolving to groundwater. However, in 2020, the mobility of As in the Guandu wetland remained limited for all three cases because of the sulfide reducing conditions (Figs. 7, 8).
Acknowledgments The authors would like to thank the National Science Council of the Republic of China, Taiwan, for financially supporting this research under Contract No. NSC98-2313-B-002-053-MY3.
References Conclusions The 1-D PHREEQC simulation results showed that the major As flow pathway was subsurface water movement in the Beitou–Guandu area, Taiwan. Regarding the spatial distribution of As, the low pH in both surface and subsurface water near the Beitou geothermal valley corresponded to the high solubility of Fe2? and As3?. In the downstream Guandu plain and Guandu wetland, the dissolved iron may have oxidized and precipitated as Fe oxyhydroxides and scavenged the aqueous As. The major retarding mechanism of As mobility in the surface flow pathway was surface adsorption by iron oxide minerals. Moreover, the major retarding mechanism of As mobility in the subsurface flow pathway was determined according to the adsorption reactions of iron oxide and iron sulfide minerals. Regarding the temporal distribution of As, the spring water of the Beitou geothermal valley continuously flowed to the downstream area, causing an increase in pH and decrease in redox potential. In the surface flow pathway, the As concentration decreased because of the sorption reactions on the surface of Fe oxydroxides. In the subsurface flow pathway, the As concentration downstream of the Guandu plain may have increased because of the reductive dissolution of As-containing Fe oxyhydroxides. The As in the Guandu wetland may have readsorbed onto goethite and was constrained by the precipitation of sulfide minerals during the period of 2007–2013. However, the geochemical reactions of As in the subsurface pathway gradually transformed into the oxidation of As-bearing sulfides; however, the mobility of As in the Guandu wetland remained limited under sulfide reducing conditions during the period of 2013–2020. According to the conducted geochemical modeling, As may be sorbed on solid minerals (Fe oxyhydroxides and iron sulfides), transported by subsurface water, and scavenged by aqueous Fe(II). Therefore, adsorption and desorption as well as oxidative and reductive reactions are the main mechanisms of As attenuation in geothermal environments.
123
Alexakis, D., & Gamvroula, D. (2014). Arsenic, chromium, and other potentially toxic elements in the rocks and sediments of Oropos-Kalamos basin, Attica, Greece. Applied and Environmental Soil Science, 2014. Article number 718534. Ball, J. W., Nordstrom, D. K., Jenne, E. A., & Vivit, D. V. (1998). Chemical analyses of hot springs, pools, geysers, and surface waters from Yellowstone National park, Wyoming, and vicinity, 1974–1975. USGS Open-File Report, 98–182. Bostick, B. C., & Fendorf, S. (2003). Arsenite sorption on troilite (FeS) and pyrite (FeS2). Geochimica et Cosmochimica Acta, 67, 909–921. Chiang, K. Y., Chen, T. Y., Lee, C. H., Lin, T. L., Wang, M. K., Jang, L. Y., & Lee, J. F. (2013). Biogeochemical reductive release of soil embedded arsenate around a crater area (Guandu) in northern Taiwan using X-ray absorption nearedge spectroscopy. Journal of Environmental Sciences, 25, 626–636. Chiang, K. Y., Lin, K. C., Lin, S. C., Chang, T. K., & Wang, M. K. (2010). Arsenic and lead (beudantite) contamination of agricultural rice soils in the Guandu Plain of northern Taiwan. Journal of Hazardous Material, 181, 1066–1071. Dzombak, D. A., & Morel, F. M. M. (1990). Surface complexation modeling: Hydrous ferric oxide. New York: Wiley. Gelhar, L. W., Welty, C., & Rehfeldt, K. R. (1992). A critical review of data on field-scale dispersion in aquifers. Water Resources Research, 28, 1955–1974. Kao, Y. H., Wang, S. W., Maji, S. K., Liu, C. W., Wang, P. L., Chang, F. J., & Liao, C. M. (2013). Hydrochemical, mineralogical and isotopic investigation of arsenic distribution and mobilization in the Guandu wetland of Taiwan. Journal of Hydrology, 498, 274–286. Kelepertsis, A., Alexakis, D., & Kita, I. (2001). Environmental geochemistry of soils and waters of Susaki area, Korinthos, Greece. Environmental Geochemistry and Health, 23, 117–135. Kinniburgh, D. G., & Cooper, D. M. (2004). Predominance and mineral stability diagrams revisited. Environmental Science Technology, 38, 3641–3648. Lie`vremont, D., Bertin, P. N., & Lett, M. C. (2009). Arsenic in contaminated waters: biogeochemical cycle, microbial metabolism and biotreatment processes. Biochimie, 91, 1229–1237. Liu, W. C., Hsu, M. H., Kuo, A. Y., & Kuo, J. T. (2001). The influence of river discharge on salinity intrusion in the Tanshui estuary, Taiwan. Journal of Coastal Resources, 17, 544–552. Maji, S. K., Kao, Y. H., & Liu, C. W. (2011). Arsenic removal from real arsenic–bearing groundwater by adsorption on iron-oxide-coated natural rock (IOCNR). Desalination, 280, 72–79.
Environ Geochem Health Manning, B. A., & Goldberg, S. (1996). Modeling competitive ad-sorption of arsenate with phosphate and molybdate on oxide minerals. Soil Science Society America Journal, 60, 121–131. Parkhurst, D. L. & Appelo, C. A. J. (1999). User’s guide to PHREEQC (version 2)—A computer program for speciation, batch-reaction, one-dimension transport, and inverse geochemical calculations. U.S. Geological Survey Water-Resources Investigations Report 99-4259. Pokrovski, G., Gout, R., Schott, J., Zotov, A., & Harrichoury, J. C. (1996). Thermodynamic properties and stoichiometry of As(III) hydroxide complexes at hydrothermal conditions. Geochimica et Cosmochimica Acta, 60, 737–749. Sigfusson, B., Gislason, S. R., & Meharg, A. A. (2011). A field and reactive transport model study of arsenic in a basaltic rock aquifer. Applied Geochemistry, 26, 553–564. Sigfusson, B., Meharg, A. A., & Gislason, S. R. (2008). Regulation of arsenic mobility on basaltic class surfaces by speciation and pH. Environmental Science Technology, 42, 8816–8821. Smedley, P. L., & Kinniburgh, D. G. (2002). A review of the source, behavior and distribution of arsenic in natural waters. Applied Geochemistry, 17, 517–568. Stumm, W., & Morgan, J. J. (1996). Aquatic chemistry: Chemical equilibria and rates in natural waters. New York: Wiley.
Syu, C. H., Jiang, P. Y., Huang, H. H., Chen, W. T., Lin, T. H., & Lee, D. Y. (2013). Arsenic sequestration in iron plaque and its effect on As uptake by rice plants grown in paddy soils with high contents of As, iron oxides, and organic matter. Soil Science and Plant Nutrition, 59, 463–471. Tsai, S. L., Singh, S., & Chen, W. (2009). Arsenic metabolism by microbes in nature and the impact on arsenic remediation. Current Opinion in Biotechnology, 20, 659–667. Wang, S. W., Liu, C. W., Lu, K. L., Chang, Y. P., & Chang, T. W. (2011). Distribution of inorganic As species in groundwater samples with the presence of Fe. Water Quality, Exposure and Health, 2, 181–192. Welch, A. H., Lico, M. S., & Hughes, J. L. (1988). Arsenic in ground water of the western United States. Ground Water, 26, 333–347. White, D. E., Hem, J. D., & Waring, G. A. (1963). Data of Geochemistry, 6th ed. M. Fleischer, (Ed). Chapter F. chemical composition of sub-surface waters. U.S. Geological Survey professional paper, 440-F. Yang, T. Y. (2011). Simulation the groundwater flow in the Guandu plain. MS thesis, Institute of Applied Geology, National Central University, Tao Yuan, Taiwan, 60p. Yoshizuka, K., Nishihama, S., & Sato, H. (2010). Analytical survey of arsenic in geothermal waters from sites in Kyushu, Japan, and a method for removing arsenic using magnetite. Environmental Geochemistry and Health, 32, 297–302.
123
