Redistribution of monosulfidic black oozes by floodwaters in a coastal acid sulfate soil floodplain.
Monosulfidic black oozes (MBO) are organic materials enriched in iron monosulfides. Thick layers of MBO accumulate in many canals that drain acid sulfate soil (ASS) landscapes (Bush et al. 2004). Laboratory experiments have demonstrated the propensity of these materials to deoxygenate water when brought into suspension (Sullivan et al. 2002). Minor amounts of MBO (e.g. 1 mg/L) can completely deoxygenate water within a few minutes, requiring very little time in suspension to react.
The recent extreme deoxygenation of the Richmond River, north-eastern New South Wales (NSW), during a 1-in-20 year flood has implicated the potential role of MBO. The Richmond River from Coraki to its outlet at Ballina (i.e. 23 km) experienced almost complete deoxygenation for several days following the February 2001 flood, causing an unprecedented fish kill. The NSW government responded to the magnitude of the fish kill by prohibiting commercial and recreational fishing for 4 months. This paper reports telling evidence of substantial MBO erosion and redistribution during this flood event. Field observations of the redistribution of substantial accumulations of MBO are described, linking for the first time the potentially significant interaction of these materials with floodwaters.
Methods and materials
Opportune field monitoring and observations of MBO accumulations were undertaken in ASS areas on the lower Richmond River floodplain, before and after the February 2001 flood event. The thickness of MBO accumulations and iron monosulfide content were measured before and after the flood in major flood mitigation drainage canals in 4 key ASS subcatchments on the lower Richmond floodplain: Tuckean Swamp, Bungawalbin Swamp, Rocky Creek, and Sandy Creek (Fig. 1). Tulau (1999) describes these catchments. To determine the thickness of the MBO layers pre and post flood, intact 1-m cores were collected from drainage canals at selected sites using a wet-sediment sampler fitted with a thin plastic liner. Debris and sediments were collected from the floodplain post-flood.
[FIGURE 1 OMITTED]
The total reduced inorganic sulfur (RIS) and acid volatile sulfur (AVS) content was measured for flood debris and flood deposited sediments that were collected during and immediately after the flood. The RIS (which includes AVS, pyrite, elemental sulfur) was quantified using a modified Cr[Cl.sub.2] reduction method (Sullivan et al. 2000). AVS was quantified on a separate sample by 9 M HCl dissolution using a field adapted diffusion-trap apparatus, following the method of Brouwer and Murphy (1995). The AVS fraction includes hydrogen sulfide and iron monosulfide minerals; however, iron monosulfides have been found to account for almost all of the AVS in river and drain sediments in our study area (Claff 2001). Elemental sulfur is rare in estuarine sediments (Berner 1970), whereas pyrite is common, and therefore, pyritic sulfur ([S.sub.P]) was estimated by subtracting the AVS content from the RIS content. The analyses were undertaken on field-wet samples to minimise sulfide oxidation and the results corrected for moisture content and reported on an oven-dry mass basis. The gravimetric moisture content ([theta]g) was determined by oven-drying subsamples at 105[degrees]C for 24 h.
Results and discussion
The erosion and re-deposition of MBO by floodwaters was obvious. Blackish-coloured organic materials were widely distributed across the floodplain with other flood debris and sediments (Figs 2, 3). MBO was deposited at the high water marks on the floodplain, such as at the top of floodgates (Fig. 2a), with lilies against fences (Fig. 2b) and amongst stands of trees (Fig. 2c). During the flood, thick islands of these materials were also dislodged by the fast-flowing drain-waters. Here, the term 'island' is used to describe large clumps of debris that were dislodged from within drains and transported with the floodwaters. One of these islands is shown in Fig. 2d, lodged against the main floodgate of the Tuckean Swamp. The blackish-organic deposits associated with the various forms of flood debris were clearly MBO from the flood mitigation drains. The flood deposits had the physical appearance and texture of MBO and contained substantial concentrations of AVS (e.g. 3% AVS) and pyritic sulfur (i.e. 1% Sp). The larger deposits were also mostly accompanied by water lilies that were growing prolifically in the drains prior to the flood.
[FIGURE 2-3 OMITTED]
As the floodwaters receded, many of the MBO islands moving with the floodwaters were left stranded at high water marks on the floodplain in many parts of the Richmond and tributaries. In the Bungawalbin subcatchment, for example, thick MBO islands were stranded on the wires and posts of stock fences (Fig. 2e). Some of the larger islands of MBO had slumped from the fence lines where they were stranded to the ground (Fig. 2f). Within days of the floodwaters receding the MBO islands had very little evidence of oxidation and contained substantial AVS concentrations (>2% AVS). However, within 2 weeks of the flooding, the stranded MBO islands left at the high water marks and other MBO deposits on the floodplain were readily oxidising, liberating acidity, and becoming encrusted with iron and sulfate (Fig. 3a-d).
The oxidation of MBO flood deposits was most evident in the Tuckean Swamp, where layers of MBO (0.02-0.1 m thick) had been deposited by the floodwaters on the banks and floodplain adjacent to the main drainage canal. These deposits had developed a thick oxidation crust of iron and sulfate (Fig. 3a, b). The iron and sulfate crust concealed the underlying MBO material (Fig. 3c). The thickness of the canal levee deposits in the Tuckean Swamp suggests they were eroded and deposited in the early stages of the flood, when flow velocities were greatest. The thickness of the MBO layer deposited on the floodplain generally diminished rapidly to a thin veneer away from drains, but was evident in many cases at great distances from drains. For example, a thin layer of MBO was still clearly evident to 300 m from the major canal in the Tuckean Swamp (Fig. 3e).
Pyrite and monosulfides were also part of the suspended sediment load of the Richmond River, downstream of the ASS areas. Traces of AVS indicative of monosulfides and pyritic sulfur were present in the silty-clay sediments deposited by floodwaters on the banks of the Richmond River (Fig. 2f), downstream of the 4 key ASS subcatchments (Fig. 2f). For example, the silty/clay deposits on tile banks of the Richmond River near Woodburn had 0.025% AVS and 0.05% Sp. AVS and [S.sub.P] was also part of the silty/clay stream bank deposits at other locations along the Lower Richmond River (Table 1). There was no AVS in the silty/clay bank deposits upstream of the ASS areas at Tatham (Table 1), indicating that MBO from ASS drains was the source of AVS in the suspended sediment load of the lower Richmond River.
The February 2001 flood occurred during the establishment of detailed transects to monitor the accumulation of MBO at selected sites. Pre and post flood comparisons (see Fig. 4) show that MBO was completely eroded from a shallow (1 m deep) drain at the Sandy Creek site (Fig. 1, site D). There was a distinct MBO layer of approximately 0.3 m depth at this monitoring site pre-flooding in February 2001 that contained up to 2% AVS (Fig. 4). Water lilies and reeds were abundant and elsewhere along the drain the MBO was 0.5-1 m thick. The floodwaters eroded the MBO almost completely from this shallow drain as clearly illustrated in Fig. 4. In the deeper drains, such as the Tuckean Swamp, floodwaters eroded only part of the MBO, with substantial amounts of MBO remaining. In these deeper drains, some of the MBO was eroded and deposited within the canal as part of the bed load. All of these observations indicate that MBO is readily mobilised and was intimately mixed and redistributed with floodwaters during the February 2001 flood.
[FIGURE 4 OMITTED]
Some 40 km of the Richmond River was severely deoxygenated with dissolved oxygen ~0.1 mg/L during the February 2001 flood. The deoxygenation of this river occurred almost at the flood peak, within a few days of the deluge (Sullivan and Bush 2001). Detailed water quality and hydrographic data for the nearby Clarence River presented by Johnston et al. (2003) also show a sharp initial drop in dissolved oxygen concentrations associated with the onset of high velocity flows in 2 floodplain drainage canals in ASS landscapes. Similar to most of the drainage canals on the Lower Richmond River floodplain, the canals examined by Johnston et al. (2003) are located in ASS and are known to accumulate MBO. The ability of MBO to react and consume oxygen very rapidly has been demonstrated in laboratory studies (Sullivan et al. 2002). The direct evidence of MBO mobilisation presented in this paper implicates MBO mobilisation as a probable cause for the initial sharp deoxygenation for floodwaters. Further studies are currently being undertaken by the authors to examine the affects of MBO on water quality.
Our observations indicate that a significant amount of MBO was eroded and redistributed during a flood event associated with extreme deoxygenation. MBO are fine-grained sediments that are readily eroded from ASS drains by mass movement and in the suspended load by floodwaters. Laboratory studies and field observations have demonstrated how MBO can react rapidly to deoxygenate water. Our field observations of MBO mobilisation provide valuable field evidence to help explain how these materials may interact and contribute to the deoxygenation of floodwaters.
Table 1. AVS and pyrite content in silty/clay river bank deposits Sampling locations are shown on Fig. 1 Site Location AVS CRS (refer to Fig. 1) (%) (%) Dungarubba A 0.004 0.09 Woodburn (downstream of B 0.025 0.05 bridge) Woodburn (at bridge) C 0.006 0.05 Sandy Creek D 0.014 0.09 Tatham (upstream of ASS E Below 0.01 areas) detection
Berner RA (1970) Sedimentary pyrite formation. American Journal of Science 268, 1-23.
Brouwer H, Murphy T (1995) Diffusion method for the determination of acid-volatile sulfides (AVS) in sediments. Environmental Toxicology and Chemistry 13, 1273-1275.
Bush RT, Fyfe D, Sullivan LA (2004) Occurrence and abundance of monosulfidic-black-ooze in a coastal acid sulfate soil landscape. Australian Journal of Soil Research 42, 609-616. doi: 10.1071/SR03077
Johnston SG, Slavich P, Sullivan LA, Hirst P (2003) Artificial drainage of floodwaters from sulfidic backswamps: effects on deoxygenation in an Australian estuary. Marine and Freshwater Research 54, 781-795. doi: 10.1071/MF02016
Claff S (2001) Formation and accumulation of sulfides in monosulfidic black drain oozes. BSc thesis, Southern Cross University, Australia.
Sullivan LA, Bush RT (2001) Report on deoxygenation and acidification of water by monosulfidic black ooze. In 'Floods and fish kills workshop'. Tropical Fruit Research Station, NSW Agriculture, Alstonville. (NSW Agriculture: Alstonville, NSW)
Sullivan LA, Bush RT, McConchie DM (2000) A modified chromium-reducible sulfur method for reduced inorganic sulfur: Optimum reaction time for acid sulfate soils. Australian Journal of Soil Research 38, 729-734.
Sullivan LA, Bush RT, Fyfe D (2002) Acid sulfate soil drain ooze: distribution, behaviour and implications for acidification and deoxygenation of waterways. In 'Acid sulfate soils in Australia and China'. (Eds C Lin, MD Melville, LA Sullivan) pp. 91-99. (Science Press: Beijing)
Tulau MJ (1999) 'Acid sulphate soil management priority areas on the Lower Richmond River floodplain.' (NSW Department of Land and Water Conservation: Sydney)
Richard T. Bush (A), Leigh A. Sullivan (A), Diane Fyfe (A), and Scott Johnston (B)
(A) Centre for Acid Sulfate Soil Research, Southern Cross University, Lismore, NSW 2480, Australia.
(B) NSW Agriculture, Grafton, NSW 2460, Australia.
Manuscript received 16 May 2003, accepted 3 May 2004
|Printer friendly Cite/link Email Feedback|
|Author:||Bush, Richard T.; Sullivan, Leigh A.; Fyfe, Diane; Johnston, Scott|
|Publication:||Australian Journal of Soil Research|
|Date:||Sep 1, 2004|
|Previous Article:||Soil properties in and around acid sulfate soil scalds in the coastal floodplains of New South Wales, Australia.|
|Next Article:||Occurrence and abundance of monosulfidic black ooze in coastal acid sulfate soil landscapes.|