|
Introduction
The Great Barrier Reef Marine Park Authority, in conjunction with other agencies, runs a multi-institutional research effort to collect quantitative and qualitative information on the composition and spatial dynamics of flood plumes. The sampling design was initially envisaged to sample flood events and at this stage, five Great Barrier Reef cyclone-initiated events and their associated flood-waters have been studied and mapped (through aerial photography) since 1994. It is worth noting that not all flood events are cyclone related as is evidenced by the recent high rainfall and discharge events in September 1998. The effects of terrestrial run-off on inshore coral reefs are, however, still only poorly understood and this research program aims to quantify some of these effects. This program aims to sample flood events as rapidly as possible after the onset of high river discharge to:
Eventually it may be possible to link land use and catchment characteristics with the composition of the plume, and to estimate the short- and long-term effects of flood waters impinging on reef biota. This paper presents the mapped distribution of the flood plumes associated with cyclone Sid (December 1997 and January 1998) and cyclone Katrina (January 1998) as well as preliminary results from the composition and evolution of nutrient concentrations and particulate matter in the freshwater flood plumes from two Queensland rivers. A summary of the weather conditions and river flow over the duration of the flood event is presented for the Russell-Mulgrave and Burdekin catchments. Methodology Over the monsoonal season, weather reports were monitored closely and all low pressure systems monitored. Heavy rain caused by Sid and Katrina in early January of 1998 was the catalyst for the flood monitoring contingency plan to be set into motion. As soon as logistically possible after the onset of flooding, aerial mapping of the flood plume in terms of spatial and temporal movement and sampling for water quality parameters commenced. Cyclone Sid and cyclone Katrina Cyclone Sid developed to the north of Arnhem Land on 26 December 1997 and moved south-east through the Gulf over the next three days. The cyclone subsequently weakened and formed a rain-bearing depression over land on 29 December, and slowly moved south for the next week. As a rain depression it caused widespread wind and rain damage in the areas between and including the Barron and Burdekin catchments. There was widespread flooding in these rivers with heavy rains falling on the upper Burdekin catchment, resulting in significant floods south of Ingham and north of Ayr. Cyclone Katrina developed in the Coral Sea on 3 January 1998 approximately 700 kilometres due east of Cairns. The cyclone intensified over the next week and this produced heavy rains along the north Queensland coast. The cyclone weakened and intensified as it travelled erratically around the Coral Sea over the next 14 days. Its movement in the Coral Sea partly contributed to the rain falling over the Queensland coast, though most flooding rivers were receiving heavy rains from ex-cyclone Sid. Plume mapping and sampling Flood plumes discharged from catchments between Ayr and Cairns following cyclone Sid and cyclone Katrina were mapped along and outwards from the Queensland coast on 13 and 14 January (figure 1). Mapping of the areal extent of the plumes was achieved by tracking the obvious brown turbid water masses contrasting with clear reef waters. The locations of the plume fronts were fixed with Geographic Positioning Systems (GPS) and loaded into a Geographic Information System (GIS) where the plume coverage was calculated. |
|
The main focus of sampling the inshore plume was to take water samples of the initial intrusion of the freshwater plume and to identify concentration gradients of water quality parameters. Salinity and temperature depth profiles were also recorded. Water samples were collected 0.5 metres below the seawater surface in clean containers at each sampling site. Water samples were collected from multiple sites from plumes originating from the Burdekin, Herbert, Tully, Johnstone, Russell-Mulgrave and Barron rivers. Water samples were collected for an initial three days at the onset of flooding, with further collections off Townsville and the Burdekin over the next two weeks. Water samples were analysed for salinity, temperature, suspended solids, chlorophyll a and phaeophytin, dissolved inorganic nutrients (ammonia, nitrate, nitrite and phosphate), dissolved organic nitrogen and phosphorus, particulate nitrogen and phosphorus. Laboratory analysis procedures are outlined in detail in Devlin and Lourey (1996). Results Results presented here summarise data collected in flood plumes adjacent to the Russell-Mulgrave River and the Burdekin River. This provides a good contrast between a wet tropic river, which floods annually, and a dry river, with a greater catchment area and more infrequent flood events from a drier catchment (table 1).
Table 1. Comparison of land use areas and other statistics for wet tropics
a. Weather characteristics
Table 2. Characteristic wind speed and direction in the Great Barrier
The Burdekin River (130 000 km2) is one of the largest catchments (other than the Fitzroy) draining into the Great Barrier Reef (Hausler 1990). Annual discharge from this river varies considerably from year to year, with major flood events separated by long, drier periods with little river flow (figure 2a). Major flooding in the Burdekin catchment can result in high discharge rates persisting over several weeks. In 1998, low salinity, turbid water was being measured for up to three weeks after initial flooding at large distances away from the river mouth (Devlin, personal observations). In contrast, annual discharges from a wet tropic river, such as the Russell-Mulgrave, are much smaller and short-lived. Figure 2b demonstrates how the Russell-Mulgrave River displays episodic flooding with one or more major flows occurring almost every year. Figure 3 presents the flow rates for the Burdekin (figure 3a) and Russell-Mulgrave rivers (figure 3b) over December and January. The dominant factor in the development of a flood plume is rainfall as this induces river run-off, which subsequently discharges its suspended load to the marine environment. Heavy rains fell on the Russell-Mulgrave and Burdekin catchments from 7 to 15 January, following persistent rainfall in late December (figure 4). In the Burdekin catchment, heavy rains fell in the upper catchment and resulted in peak flow rates from 10 to 12 January. Figure 3a shows just how quickly the Burdekin River can reach peak flow and the amount of water that can be discharged. Rainfall in the Russell-Mulgrave catchment was not as heavy as in the Burdekin area but flow did increase rapidly (figure 3b). Predominant east and south-east winds worked to constrain the plume in a northerly onshore direction. The extent of the Burdekin plume was large and was furthered by the high amount of rain that fell over Townsville (figure 4b) which resulted in large flooding of all surrounding rivers. Flooding from Townsville rivers added significantly to the extent and movement of the northward constrained Burdekin plume. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
b. Characteristics and extent of plume The plume emanating from the Russell-Mulgrave had a reduced extent and composition (figure 1) but did merge with other wet tropics plumes in a northerly direction in a very constrained nearshore band. It is worth noting that the aerial fly over and water sampling of the northern flood plume was done on 13 and 14 January, three days after peak flows, and may not be representative of the total extent of the plume and maximum concentrations. Previous work has shown (Taylor and Devlin 1997; Brodie and Furnas 1996; Devlin 1997) that the timing of sampling is critical to obtaining reliable estimates of material exported in the flood plumes. Due to this time lag, concentrations measured in the Russell-Mulgrave plume were considered to be more representative of a several day old plume mass. Salinities ranged from 2 to 28.7, with a general trend of increasing salinity away from the river mouth. The low salinity was measured close to the river mouth, with all other sites ranging from 20 to 27.8, indicating that water mixing processes have been occurring. Dissolved nutrients are generally slightly higher than ambient concentrations, with sites 1 and 2 having significantly higher concentrations of inorganic nitrogen and phosphorus. In particular, Site 2, which is north of the river mouth, has the highest concentrations of inorganic nutrients. This is most likely to be related to the northward movement of the plume and desorption of the inorganic nutrients from the particulate phase as the river water mixes with seawater (Cosser 1989).
Table 3. Range of values measured inside Russell-Mulgrave and Burdekin plumes in 1998 compared to long-term mean concentrations (*Furnas et al. 1996)
Salinities in the Burdekin plume range from 0.5 to 26.4, generally increasing northwards away from the mouth, though there were low salinities measured at sites 7 and 8, which are just north and outwards of a small flooding tributary (figure 6). Elevated levels of dissolved inorganic phosphorus were recorded at nearly all sites, with levels greater than 0.5 mM adjacent and north of the Burdekin River (figure 6). Dissolved phosphorus at the mouth is present only as dissolved inorganic phosphorus with non-detectable concentrations of dissolved organic phosphorus. Currently the calculation of particulate nutrients for this study is unavailable. However, other studies have shown that particulate phosphorus concentration is the largest concentration of all phosphorus in the river (Brodie and Mitchell 1992; Furnas et al. 1996). High phosphate (PO4) in the plume is a result of both high PO4 in the river plus desorbed PO4 from the particulate stage. Desorption of PO4 from particulate phosphorus is a commonly observed process in river plumes (Brodie and Mitchell 1992). This desorption process may allow phosphorus to remain in the water column as the plume moves offshore rather than settling near the coast via sedimentation processes. Levels of dissolved inorganic nitrogen were very high in Burdekin plume samples. The number of sites with elevated nutrients suggests clear evidence that the high nutrient composition of the Burdekin plume does carry over lengthy spatial and temporal scales. Suspended sediment levels were high with significantly elevated concentrations adjacent and north of the flooding river. Reefs and seagrass beds in the Great Barrier Reef lagoon will have experienced dissolved inorganic nitrogen concentrations greater or equal to 5 mM for periods of up to two weeks. At these levels, significant detrimental effects on the biota would be expected. The relationship between salinity and dissolved silica (figure 6e) displays typical characteristics exhibited by conservative mixing behavior. The slope of the relationship indicates that the waters are being physically mixed with little deviation from the theoretical dilution time. Given the wide expanse over which samples are collected, the extent of normal estuarine mixing appears to have moved offshore and north to Magnetic Island. Conclusion This is a very brief summary of the data that has been taken over the previous wet season. Further analysis of the samples for particulate concentrations will show a more complete picture of nutrient transport into the reef lagoon from the flooding rivers. Movement of the visibly discernible plume over the following fortnight and changes in water quality concentrations will also be presented in a later publication. This is still a major area of research and many questions are still to be answered. The subtle, indirect effects of elevated nutrients on the biota in the lagoonal system is not clear and the long-term effects of chronic stress upon the communities are largely, if not entirely, unknown. Acknowledgements Thanks to everyone who helped collect samples even in the most unpleasant weather, David Haynes for advice and support, the Department of Natural Resources for access to the flow data and the Great Barrier Reef Marine Park Authority for support of this project. References Brodie, J. and Furnas, M. 1996, Cyclones, river flood plumes and natural water quality extremes in the central Great Barrier Reef, in Downstream Effects of Land Use, eds H.M. Hunter, A.G. Eyles and G.E. Rayment, Department of Natural Resources, Brisbane, Queensland, Australia, pp. 367-374. Brodie, J. and Mitchell, A.W. 1992, Nutrient composition of the January 1991 Fitzroy River flood plume, in Workshop on the Impacts of Flooding, ed. G.T. Byron, Workshop Series No. 17, Great Barrier Reef Marine Park Authority, Townsville, pp. 56-74. Cosser, P.R. 1989, Nutrient concentration - flow relationships and loads in the South Pine River, south-eastern Queensland. I. Phosphorus loads, Australian Journal of Marine and Freshwater Research, 40: 613-630. Devlin, M. 1997, Offshore measurements late in the river plumes associated with Cyclone Sadie, in Cyclone Sadie Flood Plumes in the Great Barrier Reef Lagoon: Composition and Consequences, ed. A. Steven, Workshop Series No. 22, Great Barrier Reef Marine Park Authority, Townsville, pp. 45-53. Devlin, M.J. and Lourey, M.J. 1996, Water Quality - Field and Analytical Procedures, Long-term monitoring of the Great Barrier Reef Standard Operational Procedure No. 4, Australian Institute of Marine Science, Townsville. Furnas, M., Mitchell, A.W. and Skuza, M. 1996, Dissolved and particulate nutrients in rivers flowing into the Great Barrier Reef, Unpublished report to the Great Barrier Reef Marine Park Authority. Hausler, G. 1990, Hydrology of north Queensland coastal streams and their groundwaters, in Land Use Patterns and Nutrient Loading of the Great Barrier Reef Region, Proceedings of the workshop held at James Cook University of North Queensland, 17-18 November, 1990, ed. D. Yellowlees, Sir George Fisher Centre for Tropical Marine Studies, James Cook University of North Queensland, pp. 90-107. Steven, A., Devlin, M., Brodie, J., Baer, M. and Lourey, M. 1996, Spatial influence and composition of river plumes in the central Great Barrier Reef, in Downstream Effects of Land Use, eds H.M. Hunter, A.G. Eyles and G.E. Rayment, Department of Natural Resources, Brisbane, Queensland, Australia, pp. 85-92. Tarte, D., Hall, M. and Stocks, K. 1996, Issues in the Queensland marine environment, in The State of the Marine Environment Report for Australia Technical Annex: 3, State and Territory Issues, eds L.P. Zann and D.C. Sutton, Great Barrier Reef Marine Park Authority, Townsville, pp. 39-60. Taylor, J. and Devlin, M. 1997, The protean nature of 'wet tropical coast' flood plumes in the Great Barrier Reef Lagoon - distribution and composition, in The Great Barrier Reef Science, Use and Management, A National Conference, Proceedings, Volume 2, pp. 25-30.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Great Barrier Reef Marine Park Authority
PO Box 1379 TOWNSVILLE QLD 4810. Phone: (07) 4750 0700, Fax: (07) 4772 6093
E-mail: registry@gbrmpa.gov.au
Wolanski, E. and van Senden, D. 1983, Mixing of Burdekin flood waters in the Great Barrier Reef, Australian Journal of Marine and Freshwater Research, 34: 49-63.