Occurrence and abundance of monosulfidic black ooze in coastal acid sulfate soil landscapes.
Organic oozes enriched in iron monosulfides are called monosulfidic black ooze (MBO). The reduction of organic matter by sulfate-reducing bacteria in these sediments produces hydrogen sulfide (H2S), which reacts with and precipitates soluble iron (Berner 1970). Monosulfides are the initial sulfides to precipitate, and they form rapidly under reducing conditions in estuaries, mangrove swamps, coastal lakes, and salt marshes (Howarth and Merkel 1984; Bush and Sullivan 1997), and tidal swamps and brackish lakes (Gagnon et al. 1995). In natural marine, estuarine, and brackish environments the monosulfides occur in only small concentrations (e.g. 0.060.29% acid volatile sulfur (AVS); Morse and Cornwell 1987), progressively transforming to pyrite (Rickard et al. 1994; Gagnon et al. 1995).
Laboratory experiments have linked the oxidation of iron monosulfides contained in black organic drain ooze to rapid de-oxygenation and acidification of waters (Sullivan and Bush 2000). In this process, dissolved oxygen is consumed by the oxidation of iron monosulfide, producing ferrous iron and zero-valent sulfur. Acidification occurs after the dissolved oxygen concentration rises sufficiently to allow the acid-producing oxidation of ferrous iron to ferric iron and sulfur to sulfate. Severe de-oxygenation and acidification are often associated with flooding in the lower Richmond River, north-east New South Wales (Sammut et al. 1995; Sammut et al. 1996). The upstream extent affected by de-oxygenation in the Richmond River following the February 2001 flood coincided with the distribution of acid sulfate soil (ASS).
The objective of this paper was to examine the relationship between ASS and MBO. The distribution and abundance of MBO were quantified in drainage canals in the lower Richmond River floodplain. Iron monosulfide and pyrite concentrations in the surficial drain sediments of 3 major tributaries associated with poor water quality in the lower floodplain were measured.
Materials and methods
Drain samples were quantified in 3 subcatchments of the Richmond River: Tuckean Swamp, Rocky Mouth Creek, and Sandy Creek. The location of the subcatchments is shown in Fig. 1. Sampling locations and detail of the catchments are shown in Figs 2, 3, 4.
[FIGURE 1-4 OMITTED]
The Tuckean Swamp (Fig. 2) joins the Richmond River via the Tuckean Broadwater, a shallow brackish estuary. The swamp is a backplain occupying 5000 ha, with 3000 ha being ASS, and is part of a 22 000-ha catchment. Melaleuca quinquenervia (tea-tree) and wet meadows were the dominant vegetation prior to drainage and agricultural development (Sammut et al. 1996). A typical ASS profile consists of a peaty clay loam (0.1-0.4 m) overlying brown-grey mottled clay (>0.4 m) (Tulau 1999). Sugarcane and beef are produced in the Tuckean Swamp. A total length of 110 km of drains with an estimated drainage volume of 0.8 x 106 [m.sup.3] dissects the swamp to aid agricultural production. Many of the drains are broad and deep. For example, the Main Drain (Fig. 2) is 4.5 km long, 25 m wide, and 4-5 m deep (Sammut et al. 1996). The Bagotville Barrage (Fig. 2) separates the swamp from the tidal Tuckean Broadwater.
Rocky Mouth Creek
Rocky Mouth Creek suffers chronic and severe acidification (Tulau 1999). The catchment covers 6500 ha and has a 2400-ha floodplain/backswamp, of which 1530 ha is ASS. A typical ASS profile consists of a dark brown clay at 0-0.3 m depth, a dark grey clay with red and yellow mottles at 0.3-0.9 m, yellow mottled jarositic layer 0.9-1.1 m, overlying a dark grey clay (>1.1 m) (Tulau 1999). Sections of Rocky Mouth Creek were straightened and floodgates installed in 1965. The main floodgate located in the mid-reaches of the creek (Fig. 3) has not operated since 1994, allowing free tidal exchange.
Sandy Creek has 1350 ha of ASS in a 15 000-ha catchment. It is a tidal brackish-water creek. Flow in the creek is not regulated, but extensive areas of floodplain are drained using tidal floodgates. Several small drainage networks fitted with a l-way floodgate discharge on low tide into Sandy Creek. Bora--Codrington Drain passes through a large part of the south-eastern ASS area (Fig. 4). A typical ASS profile consists of a black peaty loam (0-0.5 m) overlying a black medium clay (> 1 m) (Tulau 1999). Tea-tree oil, sugarcane, and beef are key agricultural industries on the floodplain.
Bottom sediments were collected using a Van-veen grab from along the centerline of rivers and drains, sampling to about 0.2 m depth beneath the ooze/water interface. Intact 1-m cores were collected at selected sites using a wet-sediment sampler fitted with a thin plastic liner. Gravimetric moisture content ([theta]g) was determined at each site by oven-drying subsamples at 105[degrees]C for 24 h.
AVS and pyrite
AVS and pyritic sulfur ([S.sub.Py]) were quantified using a field-adapted diffusion-trap apparatus following the method of Hsieh and Shieh (1997). In this method, iron monosulfide minerals are dissolved by 9 M HCl within an enclosed nitrogen-purged apparatus. The evolved [H.sub.2][S.sub.(g)] precipitates as ZnS in a buffered Zn acetate solution (3% Zn acetate/2.8% ammonium hydroxide) contained in a separate vial within the apparatus. Sulfide content in the zinc acetate trapping solution was determined by iodometric titration using a 1% starch indicator. The iodine/potassium iodate titrant was standardised with 0.025 M sodium thiosulfate ([Na.sub.2][S.sub.2][O.sub.3]*5[H.sub.2]0). Using a micro-burette (i.e. 0.01-mL graduations), the practical detection limit of sulfur for this titration method is 0.001%. Pyrite is readily reduced to [H.sub.2]S by Cr[Cl.sub.2] and is quantified directly in the second step of the procedure. A new trapping solution for the [S.sub.Py] is placed in the diffusion apparatus and 20 mL of a 2 M Cr[Cl.sub.2] solution is added and allowed to stand overnight. The trapped [H.sub.2]S was quantified by titration following the procedure described for AVS. Samples were analysed in duplicate and the results corrected for gravimetric moisture content.
Total C% was determined on powdered duplicates (200 mg, finely ground samples) by combustion using a LECO 220 Sulfur/Carbon analyser. The carbonate fraction was estimated by difference in LECO C% before and after treatment with 2 M HCl, with organic carbon (OC) being the LECO C% after HCl treatment.
Acidic water quality prevailed in the ASS floodplain drains and the upper reaches of Rock Mouth Creek and Sandy Creek (Table 1). The pH of drains in the Tuckean Swamp varied from extremely acidic in the Main Drain, to slightly acidic in Tucki Drain and near neutral in the non-ASS affected Yellow Creek (Table 1, see Fig. 2 for locations). Near-neutral pH conditions also occurred in the saline/brackish tidal areas--Tuckean Broadwater, Sandy Creek, and the lower part of Rocky Mouth Creek.
The MBO occurred as a distinct layer in the bottom of the canals and rivers, with a sharp boundary between the underlying grey (Gley 6/5B), clayey sediments. The MBO was strikingly black (Gley 2.5/N), typically had moisture contents of 80-90%, and behaved like a thick fluid when handled. The MBO layer varied in thickness from 0.05 to 0.5 m at the sites examined in this study. The depth of the overlying water in the canals and rivers also varied greatly. For example, water depth in the canals that contained MBO varied from 6 to 10 m in Rocky Mouth Creek, 2 to 3 m at Tuckean Swamp, and Sandy Creek, and 1 to 1.5 m in the Bora-Codrington Drain in the Sandy Creek catchment. There was no apparent relationship between overlying water depth and MBO properties.
Abundant decaying organic matter was associated with the prolific growth of acid tolerant water lilies in acidic drains and in the upper reaches of Rocky Mouth Creek (Table 2). This resulted in dramatic increases in organic carbon contents and accumulation of black-organic oozes directly upstream of floodgates. The remains of lilies at varying stages of decay were common in the bottom sediments. Organic carbon content of sediments in the floodgated drains varied greatly, ranging from 6 to 27% (Table 2). Organic carbon was far less abundant (i.e. <2% C) in sediments from the Richmond River, Tuckean Broadwater, Sandy Creek, and the lower part of Rocky Mouth Creek. These sediments had a dark grey, rather than black, appearance.
The abundance of AVS in the 3 subcatchments is shown in Figs 5, 6, 7. In the drains and tributaries affected by ASS leachate, the concentration of AVS (i.e. iron monosulfidic sulfur) was generally an order of magnitude greater than has previously been reported for natural sediments (e.g. up to 7% in the Tuckean Swamp, Fig. 5). The Tuckean Swamp and the upper reaches of Rocky Mouth Creek illustrate the strong association of AVS, and therefore MBO, with ASS landscapes. Very little AVS occurred in the non-ASS areas like Yellow Creek, Tuckean Swamp, Sites 17 19 (Fig. 5). The AVS contents of sandy sediments of the Richmond River/Broadwater, Lower Rocky Mouth Creek, and Sandy Creek were within the range typical for estuarine sediments (i.e. 0.3%, Morse and Cornwell 1987). These results indicate that acidic products leaching from the surrounding ASS and the substantial accumulation of decaying organic matter in ASS drains are key factors contributing to the large amounts of AVS, linking ASS to MBO accumulation in coastal floodplain drains.
[FIGURE 6-7 OMITTED]
Floodgates abruptly separate the natural estuarine environments from acidic, lily-infested ASS drains. This was evidenced by the large influence that floodgates have on the abundance of AVS. Directly upstream of floodgates the AVS content increased dramatically (e.g. see Figs 7, 9). For example, at Bagotville Barrage in the Tuckean Swamp (Fig. 7), AVS content increased sharply from 0.4% in the tidal Tuckean Broadwater to 2.9% immediately upstream of the floodgate. There is also a dramatic decrease in pH upstream of the barrage (Table 1).
[FIGURE 9 OMITTED]
Pyrite was also abundant in the MBO (Figs 8, 9, 10), with [S.sub.Py] concentrations ranging up to 6%, often greatly exceeding the concentration of AVS in the bottom sediments at Rocky Mouth Creek and Sandy Creek. Spy concentrations in the Sandy Creek subcatchment ranged between 0.12 and 1.6% (Fig. 10), and at Rocky Mouth Creek followed a similar trend to AVS, dramatically increasing near and upstream of the disused floodgate (Fig. 9). The concentration of [S.sub.Py] in the upper reaches of Rocky Mouth Creek was generally more than double the concentration of AVS. At the Tuckean Swamp, although the total amount of reduced sulfur (AVS + [S.sub.Py]) in the ASS drains was similar to the acidic upper reaches of Rocky Mouth, there was proportionally far less [S.sub.Py] (Fig. 8).
[FIGURE 8 & 10 OMITTED]
Profile distribution of AVS and pyrite
The AVS and [S.sub.Py] contents of intact cores at 0.1-m depth increments in the MBO layer at 10 sites were determined. Cores were collected from the Tuckean Broadwater, the lower and upper sections of the Tuckean Swamp Main Drain, Bora-Codrington Drain, and Rocky Mouth Creek. At each site AVS was most abundant in the uppermost sediment layers. For example, at the Tuckean Swamp in the Main Drain, AVS in the upper 04).1 m layer was 3.35%, decreasing to 1% at 0.2-0.3 m depth (Fig. 11). AVS is typically more abundant near the sediment/water interface (Berner 1970; Goldhaber and Kaplan 1974). Spy at these sites was also generally most abundant in the uppermost layers.
[FIGURE 11 OMITTED]
The general correlation between Sly and AVS concentrations in the Tuckean Swamp MBO layer was reflected in the surfical MBO samples from the ASS drains in this catchment. There was a strong positive correlation between the abundance of AVS and Spy for the surficial samples of MBO from the Main Drain ([r.sup.2] = 0.78, n = 8) and Tucki drain ([r.sup.2] = 0.95, n = 6). However, this relationship was not strong and consistent for the surficial MBO samples throughout the other catchments. For instance, the abundances of AVS and Spy were only weakly correlated in the downstream section of Rocky Mouth Creek ([r.sup.2] = 0.44, n = 9; Fig. 3, Sites 1-9 for location). There was no relationship between Spy and AVS content ([r.sup.2] < 0.1) in the upper reaches (Fig. 3, Sites 10-18).
The results conclusively illustrate that AVS abundance is linked to the surrounding ASS landscape and, in particular, sections of drains where flow is gentle and affected by ASS leachate. In contrast, little AVS was present in the bottom sediments from the Richmond River, Tuckean Broadwater, lower Rocky Mouth Creek, and Sandy Creek. Tidal floodgates and drain networks in ASS landscapes provide excellent conditions for the accumulation of highly toxic and extremely reactive MBO.
Ample sulfate and iron from the surrounding ASS, combined with abundant organic matter and protected flow in ASS drains and parts of Rocky Mouth Creek, make ideal conditions for decaying organic detritus to accumulate and fuel sulfate reduction, leading to iron sulfide precipitation. The thickness of MBO and the abundance of AVS varied considerably along the ASS drains. This probably reflects the various geomorphic features of the channel bed (e.g. bars, pools, or riffles).
There is a huge reservoir of MBO with large amounts of AVS in ASS drains on the lower Richmond River floodplain. In the Tuckean Swamp alone, there is 108 km of drains. The amount of MBO in the Tuckean Swamp drains has been conservatively estimated at 200 000 [m.sup.3] (Sullivan and Bush 2000). Ooze mobilised in the Tuckean Swamp during a flood in February 2001 was capable of de-oxygenating 200 mm of runoff from the 22 000-ha catchment (nb: The catchment received 320 mm of rain; Bush et al. 2004). There are another 240 floodgated drains on the lower Richmond River floodplain.
Opening or removing floodgates to improve drainage water quality in ASS areas is being seriously considered. However, this study has shown that simply opening floodgates will not be sufficient to impede MBO accumulation. The formation of AVS in Rocky Mouth Creek shows that extreme concentrations of AVS are associated with ASS landscapes and are not restricted to floodgate controlled drains. The opening of the Rocky Mouth Creek floodgate does appear to have diffused the boundary between the lower tidal estuary and the ASS-affected sections where MBO is abundant, but has not prevented its significant occurrence. The results indicate that floodgate management on its own will not significantly limit the abundance of AVS in ASS drains.
AVS is efficiently converted to pyrite in most marine and estuarine sediments. This is evident by their typically small amounts of AVS (i.e. <0.3%; Morse and Cornwell 1987) and low AVS : Spy ratios (<1; Gagnon et al. 1995). The AVS : Spy ratios of the surficial sediments in Rocky Mouth Creek, Sandy Creek, and the Tuckean Broadwater were < 1, indicating the efficient conversion of AVS to pyrite. However, the ASS drains in the Tuckean Swamp had an average AVS : [S.sub.Py] ratio of 1.6, indicating a poor conversion of AVS to pyrite. The large differences in the proportions of AVS and [S.sub.Py] in the Tuckean ASS drains can indicate 1 of 2 differences in the sulfidisation process. Either the sediments in the Tuckean Swamp drains have accumulated more recently and are in an early phase of pyrite diagenesis, or the conversion of AVS is being limited by a factor other than time. The processes and factors affecting AVS and pyrite formation in natural estuarine and marine environments have been comprehensively described (e.g. Goldhaber and Kaplan 1974; Rickard et al. 1994; Wang and Morse 1996). However, there is little detailed information on these processes specific to the unique and dynamic conditions in ASS drain sediments. The authors are currently examining these processes.
Based on our observations, substantial MBO deposits will be common in ASS drains. Acid-tolerant algae and aquatic plants are prevalent in many ASS-affected creeks and drains (Dent 1986; Sammut et al. 1996), providing the favourable conditions for the formation of AVS and accumulation of MBO. There are thousands of ASS drains in coastal areas. Many of these will most likely contain thick deposits of highly toxic and extremely reactive MBO and represent a significant threat to the environment. The MBO has the potential to cause rapid and severe impacts to water quality if scoured by floodwaters, and therefore, drain management practices that reduce the accumulation of MBO may greatly reduce the environmental impacts from ASS.
This paper has provided the first comprehensive assessment of the occurrence and abundance of MBO in bottom sediments from drains and tributaries of a typical coastal floodplain. The abundance of MBO in ASS drains indicates that MBO may cause rapid and severe impacts to water quality when brought into suspension by floodwaters. The authors are exploring how drains can be managed to reduce MBO accumulation. Changing flow conditions, weed management, tidal water movement, drain maintenance, drain shape, and the management of adjacent lands are being investigated.
Table 1. Water pH at representative locations in the 3 subcatchments at the time of sediment sampling Location pH Tuckean Broadwater 7.50 Tuckean Swamp, drain immediately upstream of 4.81 Bagotville Barrage Tuckean Swamp, Main drain 3.41 Tuckean Swamp, Tucki drain 5.83 Tuckean Swamp, Yellow Creek in non-ASS area 6.30 Rocky Mouth Creek, lower reaches 7.42 Rocky Mouth Creek, upper reaches 3.13 Sandy Creek 7.11 Sandy Creek, Bora--Codrington drain 3.23 Table 2. Organic carbon content in surficial bottom sediments from representative drains and tributaries Location Total carbon (%) Richmond R, at Tuckean Broadwater junction 0.6 Tuckean Broadwater, near Bagotville Barrage 6.8 Tuckean Swamp, Main drain 6.6-15.9 Tuckean Swamp, Tucki drain 2.7-9.6 Rocky Mouth Ck, downstream of disused floodgate 1.6-8.0 Rocky Mouth Ck, upstream of disused floodgate 6.7-22.7 Sandy Creek 2.f1-14.4 Bora--Codrington drain, adjacent to Sandy Creek 9.0-27.0
Berner RA (1970) Sedimentary pyrite formation. American Journal of Science 268, 1-23.
Bush RT, Sullivan LA (1997) Morphology and behaviour of greigite from a Holocene sediment in eastern Australia. Australian Journal of Soil Research 35, 853-861.
Bush RT, Sullivan LA, Fyfe D, Johnston SJ (2004) Redistribution of monosulfidic black oozes by floodwaters in a coastal acid sulfate soil floodplain. Australian Journal of Soil Research 42, 603-607.
Dent D (1986) 'Acid sulphate soils: a baseline for research and development.' ILRI Publication 39. (International Institute for Land Reclamation and Improvement: Wageningen, The Netherlands)
Gagnon C, Mucci A, Pelletier E (1995) Anomalous accumulation of acid-volatile sulphides (AVS) in a coastal marine sediment, Saguenay Fjord, Canada. Geochimica et Cosmochimica Acta 59, 2663-2675. doi: 10.1016/0016-7037(95)00163-T
Goldhaber MB, Kaplan IR (1974) The sulfur cycle. In 'The sea. Vol. 5. Marine chemistry'. (Ed. ED Goldberg) pp. 569-655. (Wiley-Interscience: New York)
Howarth RW, Merkel S (1984) Pyrite formation and the measurement of sulfate reduction in salt marsh sediments. Limnology and Oceanography 29, 598-608.
Hsieh YP, Shieh YN (1997) Analysis of reduced inorganic sulfur by diffusion methods: improved apparatus and evaluation for sulfur isotopic studies. Chemical Geology 137, 255-261. doi: 10.1016/S0009-2541(96)00159-3
Morse JW, Cornwell JC (1987) Analysis and distribution of iron sulfide minerals in recent anoxic marine sediments. Marine Chemistry. 22, 55-69. doi: 10.1016/0304-4203(87)90048-X
Rickard D, Schoonen MAA, Luther GW (1994) Chemistry of iron sulfides in sedimentary environments. In 'Geochemical transformations of sedimentary sulfur'. (Eds MA Vairavamurthy, MAA Schoonen) pp. 168-193. (American Chemical Society: Washington, DC)
Sammut J, Melville MD, Callinan RB, Fraser G (1995) Estuarine acidification: the impacts on aquatic biota of draining acid sulfate soils in coastal floodplains. Australian Geographical Studies 33, 89-100.
Sammut J, White I, Melville MD (1996) Acidification of an estuarine tributary in eastern Australia due to drainage of acid sulfate soils. Marine and Freshwater Research 47, 669-684.
Sullivan LA, Bush RT (2000) The behaviour of drain sludge in acid sulfate soil areas: some implications for acidication of on Remediation and Assessment of Broadacre Acid Sulfate Soils'. Lismore, NSW. (Ed. P Slavich) pp. 43-48. (Acid Sulfate Soils Management Advisory Committee NSW)
Tulau MJ (1999) (Acid sulphate soil management priority areas on the Lower Richmond River floodplain. (NSW Department of Land and Water Conservation: Sydney, NSW)
Wang Q, Morse JW (1996) Pyrite formation under conditions approximating those in anoxic sediments. (I) Pathways and morphology. Marine Chemistry 52, 99-121. doi: 10.1016/0304-4203(95)00082-8
Richard T. Bush (A,B), Diane Fyfe (A), and Leigh A. Sullivan (A)
(A) Centre for Acid Sulfate Soil Research, Southern Cross University, Lismore, NSW 2480, Australia.
(B) Corresponding author; email: firstname.lastname@example.org
Manuscript received 16May 2003, accepted 20 April 2004
|Printer friendly Cite/link Email Feedback|
|Author:||Bush, Richard T.; Fyfe, Diane; Sullivan, Leigh A.|
|Publication:||Australian Journal of Soil Research|
|Date:||Sep 1, 2004|
|Previous Article:||Redistribution of monosulfidic black oozes by floodwaters in a coastal acid sulfate soil floodplain.|
|Next Article:||Occurrence of marcasite in an organic-rich Holocene estuarine mud.|