Linking Groundwater Discharge to Severe Estuarine Acidification

Mar 3, 2011 - Periodic acidification of waterways adjacent to coastal acid sulfate soils (CASS) is a significant land and water management issue in th...
0 downloads 4 Views 2MB Size
ARTICLE pubs.acs.org/est

Linking Groundwater Discharge to Severe Estuarine Acidification during a Flood in a Modified Wetland Jason de Weys,* Isaac R. Santos, and Bradley D. Eyre Centre for Coastal Biogeochemistry, School of Environmental Science and Management, Southern Cross University, P.O. Box 157, Lismore, New South Wales 2480, Australia ABSTRACT: Periodic acidification of waterways adjacent to coastal acid sulfate soils (CASS) is a significant land and water management issue in the subtropics. In this study, we use 5-months of continuous radon (222Rn, a natural groundwater tracer) observations to link estuarine acidification to groundwater discharge in an Australian CASS catchment (Tuckean Swamp). The radon time series began in the dry season, when radon activities were low (2-3 dpm L-1), and the pH of surface water was 6.4. We captured a major rain event (213 mm on 2 March 2010) that flooded the catchment. An immediate drop in pH during the flood may be attributed to surface water interactions with soil products. During the post-flood stage, increased radon activities (up to 19.3 dpm L-1) and floodplain groundwater discharge rates (up to 2.01 m3 s-1, equivalent to 19% of total runoff) coincided with low pH (3.77). Another spike in radon activities (13.2 dpm L-1) coincided with the lowest recorded surface water pH (3.62) after 72 mm of rain between 17 and 20 April 2010. About 80% of catchment acid exports occurred when the estuary was dominated by groundwater discharging from highly permeable CASS during the flood recession.

’ INTRODUCTION The impacts of coastal acid sulfate soils (CASS) are a significant land and water management issue in several subtropical countries.1 Many of Australia’s coastal floodplains have an extensive network of constructed drains, floodgates, and modified water courses. These drains are designed to mitigate the impacts of floods and have allowed agriculture and settlements to be established. Drainage of coastal floodplains lowers the water table, oxidizing iron-sulfide minerals that may result in an extremely acidic soil profile.1 The soil acidity can be exported into adjacent waterways, resulting in surface water pH as low as 3, and concentration of metals exceeding water quality guidelines.2 While groundwater is suspected to be the major source of acidity in many cases,3 no studies have demonstrated the relative contribution of groundwater discharge to surface water acidification in CASS catchments. Radon (222Rn) had been utilized as a groundwater tracer for decades as it is naturally enriched by 2-4 orders of magnitude in groundwater relative to surface water and is chemically conservative.4 Its half-life (t1/2 = 3.8 days) is comparable to many physical processes in surface waters. Early studies using 222Rn to trace groundwater were constrained by the need to collect grab samples that have to be analyzed in the laboratory. The development of automated radon monitors5 has led to a surge in the use of radon as a groundwater tracer. Several short-term (hours to days) continuous radon time series measurements have been performed in estuaries and coastal waters. However, there have been only three studies attempting to resolve radon trends over r 2011 American Chemical Society

time scales longer than one week or so.6 These investigations were focused in coastal-marine environments. For example, at a coastal site in Korea, nearly 2 months of radon data revealed significant fluctuations in concentrations.6b In freshwater systems, several studies relied on spatial radon surveys to assess groundwater inputs.7 However, there have been no long-term, high resolution radon observations in inland freshwater or estuarine systems. The aim of this study was to use long-term, continuous radon observations to reveal a link between estuarine acidification and groundwater discharge into an artificial drainage network in an Australian CASS catchment. We demonstrated the use of a passive approach to monitor radon concentrations, quantifying groundwater discharge into surface waters during contrasting hydrological conditions, and estimated the relative contribution of groundwater to acid exports from the wetland. These observations build on a previous high resolution spatial survey and 24-h time series measurements performed at the same site.8

’ EXPERIMENTAL SECTION Study Site. The Tuckean Broadwater is a tidal estuarine embayment connecting the Tuckean Swamp to the Richmond Received: December 6, 2010 Accepted: February 11, 2011 Revised: February 10, 2011 Published: March 03, 2011 3310

dx.doi.org/10.1021/es104071r | Environ. Sci. Technol. 2011, 45, 3310–3316

Environmental Science & Technology

ARTICLE

Figure 1. The Tuckean Swamp catchment in Eastern Australia. The dark areas represent waterways, while the gray scale represents the depth of coastal acid sulfate soils. The surface water time series site is indicated by the white square in the Tuckean Broadwater, while the groundwater samples are indicated by black circles. The Tuckean Broadwater is tidal and drains into the Richmond River estuary.

River (Figure 1). The Tuckean Swamp, a former large back barrier lagoon, consists of approximately 4,000 ha of CASS floodplain, with a low relief (∼1-2 m Australian Height Datum [AHD]).9 The total catchment area feeding the tributary is approximately 22,000 ha and includes the hills of the Alstonville Plateau. Over 75% of the Swamp contains partially oxidized sulfidic sediments.10 The Swamp contains a permanently high water table of one meter or less below the surface and over 110 km of drains.11 A barrage (Bagotville Barrage) limits tidal ingress into upstream agricultural areas.9 The Barrage comprises a 300 m levee with 8  2.6 m2 flap-gates that open downstream, allowing floodwaters downstream and preventing tidal flow upstream.10 The construction of the Barrage has resulted in the Swamp being transformed into a freshwater system and removed the potential for seawater to buffer the effects of acid sulfate runoff. Radon Measurements. Surface water 222Rn was measured at one hour intervals from 6 January 2010 to 1 June 2010 using a RAD7 automated radon-in-air monitor adapted for radon-inwater.5a The RAD7 was fixed on the river bank and connected to a device to extract radon from water. Two devices were trialled on three occasions over 24 h periods: (A) an air-water exchanger12 that responds quickly (∼30 min) to changes in concentrations but requires pumping water at about 3 L/min (air was pumped at 1 L/min). This system consumes one standard 12 V car battery every 12 h and (B) a hydrophobic membrane13 that responds slowly (4-6 h) to changes in concentration but does not require pumping water and thus one car battery lasts 3 weeks. The membrane was mounted on an open wire frame, suspended 40 cm below the water surface, and connected to a RAD7 with tubing supplied by the manufacturer. Radon-in-

Figure 2. Comparison of an air-water exchanger and a hydrophobic membrane used to extract radon from water when connected to a RAD7 radon-in-air monitor: (A) air-water exchanger; (B) hydrophobic membrane; and (C) an example of a 24 h time series intercomparison experiment. The exchanger detected a sudden change in radon activity after 15 h, while the slower-responding membrane did not capture this change; (D) correlations between all experiments comparing the exchanger and membrane. The deviation from the 1:1 line is within the calibration uncertainties of the RAD7s used during these experiments ((5%). When radon activities are averaged over 6-24 h periods, the two systems gave identical results.

water diffused through the membrane into the RAD7’s closed air loop until the radon concentration in air was in equilibrium with water. We used a 2 m long membrane coil (Figure 2) but different configurations may yield better response times.13a Intercomparison experiments indicated that when averaged over 6 to 24 h, these extraction devices produce indistinguishable results (Figure 2). Previous studies utilizing the RAD7 relied on the air-water exchanger.4,6c,14 However, the exchanger was not suitable for our long-term deployment as we are dealing with a remote location. The need to pump water continuously requires a reliable power source that was unavailable. The hydrophobic membrane overcomes this limitation as it could be left unattended for 2-3 weeks. This is a major advantage provided that the slower response time (4-6 h) of the membrane is acceptable. Other Measurements. A submersible multiparameter water quality data logger recorded hourly measurements of pH and 3311

dx.doi.org/10.1021/es104071r |Environ. Sci. Technol. 2011, 45, 3310–3316

Environmental Science & Technology

ARTICLE

Table 1. Summary of Groundwater Samples Obtained from the Tuckean Swamp uncertainty site no.

latitude/

radon (dpm/L)

(dpm/L)

depth (meters)

pH

sp cond (mS/cm)

temp (° C)

DO (mg/L)

longitude

Site 1

119

24

1.5

4.3

0.961

21.96

1.94

S28°580 41.500 E153°240 19.900

Site 2

410

16

2.7

6.4

0.192

23.23

2.72

S28°540 37.600 E153°250 37.100

Site 3

682

72

6.3

5.9

0.275

21.51

1.55

S28°570 47.100 E153°210 53.500

Site 4a

141

17

3.9

4.5

1.37

23.13

1.37

S28°550 16.000 E153°220 20.900

Site 4b

572

52

7.0

6.5

3.97

23.33

0.92

S28°550 16.000 E153°220 20.900

Site 5a

61

10

5.3

5.3

0.151

22.58

2.01

S28°560 50.100 E153°220 03.700

Site 5b Site 5c

118 72

25 20

2.2 7.2

4.3 7.0

0.127 0.597

23.69 21.81

1.68 1.17

S28°560 50.100 E153°220 03.700 S28°560 50.100 E153°220 03.700

Site 6

191

41

6.4

5.7

6.8

21.46

1.25

S28°590 04.700 E153°240 13.800

Site 7

194

26

1.0

3.1

1.3

23.52

3.21

S28°560 11.300 E153°240 21.900

Site 8a

130

19

1.8

3.7

2.31

23.15

3.91

S28°560 22.500 E153°230 57.600

Site 8b

923

125

12.4

6.9

3.56

20.99

1.76

S28°560 27.200 E153°230 57.400

Site 8c

139

33

2.0

3.4

2.91

22.75

1.45

S28°560 21.500 E153°230 57.700

Site 9a

118

16

1.0

3.7

1.58

22.7

1.58

S28°560 19.800 E153°240 00.600

Site 9b Site 10a

191 583

13 23

1.0 4.0

3.7 6.1

1.396 1.348

-

-

S28°560 20.100 E153°240 01.200 S28°550 42.900 E153°240 34.300

Site 10b

214

43

2.1

3.6

0.765

-

-

S28°550 42.900 E153°240 34.300

Site 11a

138

21

5.9

6.1

0.22

20.56

1.54

S28°580 36.700 E153°220 12.900

Site 11b

143

18

9.4

6.3

0.513

20.05

0.95

S28°580 36.700 E153°220 12.900

Site 12

198

28

9.1

5.6

0.11

21.67

1.23

S28°550 51.800 E153°220 15.200

Site 13a

217

58

5.8

4.9

1.8

22.32

1.41

S28°550 45.500 E153°220 38.800

Site 14

174

19

3.6

5.0

1.125

22.94

1.23

S28°560 14.900 E153°220 54.100

Site 14a

92

77

1.5

4.5

4.2

21.64

1.32

S28°560 12.400 E153°220 52.100

specific conductivity in surface waters. Continuous measurements (10 min time steps) of water level and temperature were recorded with a Van Essen datalogger fixed to the river bed. Surface water velocities were continuously monitored using a Starflow ultrasonic data logger (15 min time steps). A crosssectional area was measured so that the current meter velocities could be converted to surface water discharge. Corrections to the cross-section area were made as a function of water level fluctuations. A weather station (Davis, Vantage Pro2) was deployed within 100 m of the radon detector to record rainfall and wind speed. Averages were calculated over 24 h for each parameter sampled. A 1.5 m deep monitoring well was established 200 m upstream of the time series station. Specific conductivity, temperature, and water table height were continuously measured with a Van Essen CTD Diver. To characterize the groundwater end-member discharging to the Tuckean Broadwater, groundwater was sampled from 23 wells. Pre-existing NSW Office of Water monitoring wells were utilized as well as four new shallow wells (10%). The pH of surface waters with high radon concentrations ranged between 3.5 and 4.1, a value comparable to the lowest pH of groundwater (Table 1). This suggests that the pH of surface water is not altered after radon concentrations and the groundwater contribution reach a threshold of about 12 dpm L-1 and 10%, respectively. The removal of eleven dates (circled outliers in Figure 4) that correspond with heavy rains in the Swamp increases the correlation coefficient to r2=0.77. A decrease in pH was observed immediately following the flood and before the increase in the groundwater flux. The relatively slow response of the membrane (4 to 6 h) to changes in radon concentration cannot explain these results as the outliers were observed for up to 5 days after large rain events. There are three possible sources of low radon and low pH associated with the eleven outliers shown in Figure 4: (A) The ‘first flush phenomenon’ results in the mobilization of salts from the upper soil profile.1 The short residence time of the flood waters in the upper soil profile in contrast to the ingrowth time of radon results in low radon values. (B) The remobilization of monosulfidic black ooze (MBO). Drains in the Tuckean Swamp contain vast amounts of MBO.17 These deposits can be scoured by the strong currents created by floodwaters and release acid products.1 (C) Overland sheet flow, which is the exporting of water from already inundated areas of very low relief. It was observed that large sections of the swamp remained inundated for long periods even after only moderate rainfall. Previous studies have shown that acidic surface water inundated 40% of the swamp after a flood event. After periods of heavy rain, when overland sheet flow occurs, low pH water from already inundated areas is exported into drains.10 Additional insights into the relative contribution of these sources could likely be obtained from stable isotope tracers such as 18O and 2H. Acid Exports. Acid exports from the Tuckean Swamp were calculated by simply multiplying the Hþ concentration by the surface water discharge, assuming that all dissociated Hþ is from H2SO4 as demonstrated by earlier work.10 The export of acid was primarily controlled by the pH of surface waters as implied from stronger correlation coefficients with the Hþ concentration (r = 0.79; n = 116 days; p < 0.01) than with the surface water 3314

dx.doi.org/10.1021/es104071r |Environ. Sci. Technol. 2011, 45, 3310–3316

Environmental Science & Technology discharge (r2 = 0.14; n = 118 days; p > 0.01). The highest daily average acid flux occurred during the postflood stage, which was also the stage with the highest groundwater flux and radon activity. Toward the end of the post-flood period, the acid flux approached zero, coinciding with a drop in the groundwater flux. Approximately seven days into the minor rains stage, radon activity and the groundwater flux experienced a spike, with an immediate corresponding increase in acid flux, providing further evidence of groundwater being the main source of acidity. The estimated production of acid being exported to the estuary was 332 kg H2SO4 ha-1 year-1. About 80% of this export occurred during the post-flood and minor rains stages when groundwater clearly dominated the system, while 20% occurred during the eleven dates immediately after rains when surface water dominated the system. Observations in the 1990s yielded estimates of ∼300 kg H2SO4 ha-1 year-1 exported from the Tuckean Swamp.10 A series of management approaches have been undertaken to reduce the impacts of acid discharge in the past decade. The similar export rates found in our study compared to previous estimates implies that acid exports have not decreased since the 1990s. However, it must be taken into account that this previous study used a water balance approach, rather than direct current observations to estimate catchment water discharges. Estimates of acid production from other CASS sites in Australia have shown similar rates of production. For example, another artificially drained floodplain, located approximately 75 km north of the Tuckean Swamp, has acid production estimated to be 276 kg ha-1 year-1.1 The Tuckean Swamp contains about 1.3  106 tonnes of sulfuric acid.10 Using this estimate and assuming no further oxidization of pyrite is occurring, the Swamp will continue exporting acid products to waterways for the next 1000 years. Implications. Commonly used methods for quantifying groundwater discharge such as flow equations and hydrograph separation techniques can generate uncertainties comparable to the estimated discharge.18 This can be especially important in highly heterogeneous CASS catchments where hydraulic conductivities range over 4 orders of magnitude over small areas.3 While not without limitations19 combining the use of a natural tracer with a mass balance approach may reduce uncertainty when estimating the contribution of groundwater discharge to hydrological budgets. This information enables managers to make more informed decisions regarding the protection of environmental flows.18b Our observations provided insights into the drivers of low pH conditions in the Tuckean Swamp during a flood event. The hydrophobic membrane can be a very useful tool when it is acceptable to average radon concentrations over several hours, as it would be in most river systems. The significant correlation between radon and pH, combined with the sustained low pH conditions experienced for several weeks, indicates that groundwater is the main source of acidity in the Tuckean Broadwater. As groundwater can be a major driver of surface water quality, management decisions in other river and CASS sites would benefit from similar studies. The advent of portable radon-in-air monitors adapted to radon-in-water is allowing rapid assessments of groundwater discharge and decreasing uncertainty in hydrologic budgets.

’ AUTHOR INFORMATION Corresponding Author

*Phone: 61 746320752. E-mail: [email protected].

ARTICLE

’ ACKNOWLEDGMENT This project was supported by a Hermon Slade Foundation grant (09-01). B.D.E. and I.R.S. acknowledge the support of ARC grants (LP100200731 and DP110103638). We thank Bruce Heynatz for allowing the deployment of the radon system on his property and invaluable guidance, and Maria Matthes for allowing installation of a piezometer on her property. Ben Shepherd and Douglas Tait assisted with field work, the Richmond River County Council supplied water quality information, and the NSW Office of Water allowed access to monitoring wells. We thank the editor and four anonymous reviewers for valuable feedback. ’ REFERENCES (1) Macdonald, B. C. T.; White, I.; Astr€om, M. E.; Keene, A. F.; Melville, M. D.; Reynolds, J. K. Discharge of weathering products from acid sulfate soils after a rainfall event, Tweed River, eastern Australia. Appl. Geochem. 2007, 22 (12), 2695–2705. (2) Ferguson, A.; Eyre, B. Behaviour of aluminium and iron in acid runoff from acid sulphate soils in the lower Richmond River catchment. J. Aust. Geol. Geophys. 1999, 17 (5/6), 193–201. (3) Johnston, S. G.; Hirst, P.; Slavich, P. G.; Bush, R. T.; Aaso, T. Saturated hydraulic conductivity of sulfuric horizons in coastal floodplain acid sulfate soils: Variability and implications. Geoderma 2009, 151 (3-4), 387–394. (4) Burnett, W. C.; Peterson, R. N.; Santos, I. R.; Hicks, R. W. Use of automated radon measurements for rapid assessment of groundwater flow into Florida streams. J. Hydrol. 2010, 380 (3-4), 298–304. (5) (a) Burnett, W.; Kim, G.; Lane-Smith, D. A continuous monitor for assessment of 222Rn in the costal ocean. J. Radioanal. Nucl. Chem. 2001, 249 (1), 167–172. (b) Schubert, M.; Barkin, W.; Peza, P.; Lopez, A.; Balcazar, M. On-site determination of the radon concentration in water samples: methodical background and results from laboratory studies and a field-scale test. Radiat. Meas. 2006, 41, 492–497. (6) (a) Dulaiova, H.; Camilli, R.; Henderson, P.; Charette, M. Coupled radon, methane and nitrate sensors for large-scale assessment of groundwater discharge and non-point source pollution to coastal waters. J. Environ. Radioact. 2010, 101 (7), 553–563. (b) Kim, G.; Hwang, D. Tidal pumping of groundwater into the coastal ocean revealed from submarine 222Rn and CH4 monitoring. Geophys. Res. Lett. 2002, 29, 1678. (c) Santos, I. R.; Dimova, N.; Peterson, R. N.; Mwashote, B.; Chanton, J.; Burnett, W. C. Extended time series measurements of submarine groundwater discharge tracers (Rn-222 and CH4) at a coastal site in Florida. Mar. Chem. 2009, 113 (1-2), 137–147. (7) Cook, P.; Wood, C.; White, T.; Simmons, C.; Fass, T.; Brunner, P. Groundwater inflow to a shallow, poorly-mixed wetland estimated from a mass balance of radon. J. Hydrol. 2008, 354 (1-4), 213–226. (8) Santos, I. R.; Eyre, B. D. Radon tracing of groundwater discharge into an Australian estuary surrounded by coastal acid sulphate soils. J. Hydrol. 2011, 396, 246–257. (9) Taffs, K. H.; Farago, L. J.; Heijnis, H.; Jacobsen, G. A diatombased Holocene record of human impact from a coastal environment: Tuckean Swamp, eastern Australia. J. Paleolimnol. 2008, 39 (1), 71–82. (10) Sammut, J.; White, I.; Melville, M. D. Acidification of an estuarine tributary in eastern Australia due to drainage of acid sulfate soils. Mar. Freshwater Res. 1996, 47, 669–684. (11) Bush, R. T.; Fyfe, D.; Sullivan, L. A. Occurrence and abundance of monosulfidic black ooze in coastal acid sulfate soil landscapes. Aust. J. Soil Res. 2004, 42 (6), 609–616. (12) Dimova, N.; Burnett, W.; Lane-Smith, D. Improved Automated Analysis of Radon (222Rn) and Thoron (220Rn) in Natural Waters. Environ. Sci. Technol. 2009, 43 (22), 8599–8603. (13) (a) Schubert, M.; Schmidt, A.; Paschke, A.; Lopez, A.; Balcazar, M. In situ determination of radon in surface water bodies by means of a hydrophobic membrane tubing. Radiat. Meas. 2008, 43 (1), 111–120. (b) Surbeck, H. In A radon-in-water monitor based on fast gas transfer membranes, International conference on technologically enhanced 3315

dx.doi.org/10.1021/es104071r |Environ. Sci. Technol. 2011, 45, 3310–3316

Environmental Science & Technology

ARTICLE

natural radioactivity (TENR) caused by non-uranium mining, Szczyrk, Poland, Szczyrk, Poland, 1996; p 12. (14) (a) Peterson, R. N.; Santos, I. R.; Burnett, W. C. Evaluating groundwater discharge to tidal rivers based on a Rn-222 time-series approach. Estuarine, Coastal Shelf Sci. 2010, 86 (2), 165–178. (b) Swarzenski, P. W.; Simonds, F. W.; Paulson, A. J.; Kruse, S.; Reich, C. Geochemical and geophysical examination of submarine groundwater discharge and associated nutrient loading estimates into Lynch Cove, Hood Canal, WA. Environ. Sci. Technol. 2007, 41, 7022–7029. (15) Dulaiova, H.; Gonneea, M. E.; Henderson, P. B.; Charette, M. A. Geochemical and physical sources of radon variation in a subterranean estuary - Implications for groundwater radon activities in submarine groundwater discharge studies. Mar. Chem. 2008, 110 (1-2), 120–127. (16) (a) Eyre, B.; Twigg, C. Nutrient behaviour during post-flood recovery of the Richmond River estuary northern NSW, Australia. Estuarine, Coastal Shelf Sci. 1997, 44, 311–326. (b) Eyre, B. D.; Ferguson, A.J.P. Impact of a flood event on benthic and pelagic coupling in a subtropical east Australian estuary (Brunswick). Estuarine, Coastal Shelf Sci. 2006, 66, 111–122. (17) Eyre, B.; Kerr, G.; Sullivan, L. Deoxygenation potential of the Richmond River Estuary floodplain, northern NSW, Australia. River Res. Appl. 2006, 22 (9), 981–992. (18) (a) Burnett, W. C.; Aggarwal, P. K.; Aureli, A.; Bokuniewicz, H.; Cable, J. E.; Charette, M. A.; Kontar, E.; Krupa, S.; Kulkarni, K. M.; Loveless, A.; Moore, W. S.; Oberdorfer, J. A.; Oliveira, J.; Ozyurt, N.; Povinec, P.; Privitera, A. M. G.; Rajar, R.; Ramassur, R. T.; Scholten, J.; Stieglitz, T.; Taniguchi, M.; Turner, J. V. Quantifying submarine groundwater discharge in the coastal zone via multiple methods. Sci. Total Environ. 2006, 367 (2-3), 498–543. (b) Cook, P.; Favreau, G.; Dighton, J.; Tickell, S. Determining natural groundwater influx to a tropical river using radon, chlorofluorocarbons and ionic environmental tracers. J. Hydrol. 2003, 277 (1-2), 74–88. (19) Burnett, W. C.; Santos, I. R.; Weinstein, Y.; Swarzenski, P. W.; Herut, B. Renmaining uncertainties in the use of Rn-222 as a quantitative tracer of submarine groundwater discharge. In A new focus on groundwater-seawater interactions; Sanford, W., Langevin, C., Polemio, M., Povinec, P., Eds.; IAHS Press: Perugia, 2007; pp 109-118.

3316

dx.doi.org/10.1021/es104071r |Environ. Sci. Technol. 2011, 45, 3310–3316