Brodie et al., 2003:

"The Great Barrier Reef World Heritage Area (GBRWHA) extends along the Queensland coast for 2000 km.The coast adjoining the GBRWHA has a diverse range of wet and dry tropical catchments, covering an area of 423,000 km². Most catchments are small, but two, the Burdekin and Fitzroy, are among the largest in Australia. Land-use on the Great Barrier Reef Catchment Area (GBRCA) is dominated by rangeland beef grazing and cropping, largely sugarcane cultivation, but also horticulture and cotton, with relatively minor urban development. Exports of terrestrial sediments, nutrients and pesticide residues to the Great Barrier Reef (GBR) lagoon largely occur during periods of flood runoff.

Research and monitoring have clearly established that changed land use activity on the catchment of the GBRWHA is directly contributing to an increase in the load of pollutants discharging to the area. Many of the pollutants measurable in river flows are known to be degrading the ecosystems of the GBRWHA, particularly those close to the coast. Similar patterns of pollutant-related decline have lead to the collapse of coral reef systems and other coastal marine ecosystems in other parts of the world. Many pollutant loads continue to increase and show no sign of reduction. Of particular concern is the increase in fertilizer sourced inorganic nitrogen that is fully bioavailable to marine ecosystems.

The pollutant contributions of individual catchments of the GBRCA vary greatly. This is due to the catchment size, volume of runoff, and the land uses, soil types, vegetation cover and terrain producing the pollutant. Rivers draining into the GBR have been greatly modified since development began after about 1850. Large areas of woodland and forest have been cleared or thinned for grazing and cropping. Human populations have increased rapidly since 1950 with associated increases in sewage, urban stormwater and industrial discharges. Fertilizer use has increased rapidly. Even in recent times large expansion of agricultural activity has continued to occur.

A number of models for estimating sediment, nitrogen (N) and phosphorus (P) discharge to the GBR have been developed over the last 20 years. The sophistication and reliability of the models and data supporting them has increased through time. Sediment inputs to the GBR have been calculated from the estimated accumulation of sediment in the coastal zone or weighted discharge-export relationships derived from a small number of rivers. Simple models of run-off, land-use, and sediment delivery suggest that sediment and nutrient fluxes to the GBRWHA have increased several-fold since the commencement of European agricultural practices. Catchments with high levels of land clearing and land use, typically for beef grazing and/or fertilized cropping, show the greatest increases in sediment and nutrient export.

To begin to manage sediment and nutrient exports it is essential to identify the sources of sediment and nutrient that are exported to the coast. This is quite a different problem to mapping soil erosion and nutrient loss in the contributing catchments. A recent assessment of erosion and river sediment loads estimated that approximately 14 Mt of sediment is exported to the GBR coast each year. This is less than 10% of the total amount of erosion in the catchment. The rest is deposited on footslopes, floodplains, riverbeds and in reservoirs and does not reach the coast. Nutrients are deposited with this sediment but the bulk of dissolved nutrients are delivered to the coast.

The present project aimed to use the models, SedNet and its component nutrient model ANNEX, aspects of which were either calibrated or tested using water quality data from the GBRCA, to prioritise catchments and sub-catchments draining to the GBRWHA through identifying the sources of such material for the purpose of reducing sediment and nutrient delivery to the coast.

The sources that contribute to export at the coast were identified by modelling spatially distributed sediment and nutrient budgets in the contributing catchments. These budgets map the sources, sinks and transport of sediment and nutrient river link by river link, as it is transported to the coast. The models uses empirical and conceptual models of each of the transport processes as a function of the principal driving environmental factors. Environmental factors incorporated include land use, land cover, soil properties including nutrient content, terrain, rainfall, dissolved nutrient concentrations in runoff from differing land uses and river hydrology. Processes which trap sediments and nutrient in the catchment such as sedimentation and denitrification are also included. The guiding principle behind these budgets is that in the load of material carried by any stretch of river is determined by the rate of supply from various erosion processes and nutrient sources on the land, less losses to deposition and nutrient transformation during transport. This balance of inputs and losses is calculated in each river link sequentially from the source streams to the mouth of the catchment, gradually accumulating load. Mapping of sediment and nutrient sources, including soil erosion, gully erosion, riverbank erosion, point sources and diffuse sources in runoff is an essential input into the material budgets.

Higher resolution data of better quality is now available for catchments of the GBRWHA than was used in previous modelling and estimates and improvements were made to the model by using the understanding gained from river water quality monitoring and studies of sediment and nutrient losses from land uses in the catchments. The development of catchment modelling tools by CSIRO (the SedNet model) and the analysis of water quality water monitoring data that has been completed by Australian Institute of Marine Science (AIMS), Department of Natural Resources and Mines (DNRM) and Queensland Environmental Protection Agency (EPA) made it possible to combine all available information in the project. It also made it possible to move from mapping sediment and nutrient loss per se to mapping hot spots of increased delivery of that material to the marine environment. The modelling allows extrapolation into catchments and sub-catchments for which there is no water quality monitoring data, and allows us to predict where remedial measures will have the greatest ability to reduce future loads. These can be used to identify catchment and sub-catchments and specific land uses worthy of prioritisation for catchment restoration associated with the Reef Protection Plan, the National Action Plan for Salinity and Water Quality and the Natural Heritage Trust.

The modelling shows that soil erosion is the dominant process supplying 63 % of sediment to the rivers. Gully erosion is also a significant source of sediment, but is localised within a few catchments. At the scale of the GBR catchment, river bank erosion is relatively minor, but again, can dominate in some river basins, such as the Mary River, and is locally significant in others, such as parts of the Fitzroy and Burdekin R. basins. While all erosion processes need to be managed to reduce sediment loads in the catchment, it is clear that overall soil erosion dominates, and soil erosion control will have the greatest impact on reducing sediment delivery to rivers.

Overall, 70% of sediment contributed to the coast comes from just 20% of the total catchment area. The areas of high contribution are all relatively close to the coast. They are concentrated in the Mackay-Whitsunday catchments, Bowen River sub-catchment of the Burdekin, the Wet Tropics, the SE part of the Normanby River basin, near coast parts of the Fitzroy and Burnett River basins and the main stem of the Mary River. Some inland areas of the Fitzroy, upper Burdekin and Burnett River basins, including areas of high gully erosion are moderate contributors of sediment to the coast. Other inland areas have either low erosion rates or low delivery potential. If the goal is to reduce sediment loads to the coast then remedial works can be focussed on particular sediment sources and the land uses and erosion processes found there. Targeting the areas with a disproportionately high level of contribution and large difference to natural contribution should be a priority as this will have the greatest effect on reducing sediment export to the coast.

Each of the sediment sources described above, together with dissolved contributions from surface and sub-surface runoff deliver nutrients to the streams and rivers in the GBR catchments. Hillslope (soil) erosion is by far the largest source of particulate nutrients because of its dominance as a sediment source and the higher nutrient concentrations on surface soils. Gully and riverbank erosion make up less than 10% of the total nutrient sources. Our results predict that about 30% of N sources comes from dissolved forms in runoff, and about 15% of P is derived from dissolved sources.

The spatial patterns of total N and P contribution to streams largely reflect the soil erosion predictions so that the Mackay-Whitsunday coast, and near-coast parts of the Fitzroy River and Burdekin River basins are significant nutrient sources. In addition, the Wet Tropics is an area of high nutrient source despite only moderate soil erosion because of high dissolved losses, from both rainforest and cropland areas. Inland areas have relatively low nutrient supply to streams because of low soil erosion, relatively low nutrient concentrations and little diffuse runoff.

The ratio of current inputs to natural N and P inputs show that much of the Wet Tropics and Cape York have low increases in nutrient input because of low intensity land use. There are small areas of high increase in the intensively used lowlands of the Wet Tropics. Most of the areas of high increase (>10 times natural inputs) are in isolated areas of grazing and cropping of relatively low natural input in the Burdekin, Fitzroy and Burnett River basins. Most of these catchments, and the adjoining coastal areas have a moderate increase in nutrient input. Examining area specific total nitrogen (TN) and total phosphorous (TP) loads shows significant diffuse inputs across the Mackay- Whitsunday and some of the Wet Tropic river basins. The Burdekin and Fitzroy River basins have low specific TN and TP exports, because of the very large catchment size with extensive areas of low erosion and areas of significant deposition. For much of the area south of the Herbert River catchment current diffuse TN and TP exports are at least 5 times the predicted natural rates.

As part of this project, we ran the budget models for three possible future land use scenarios. The main purpose of these scenarios was to assess what the sediment and nutrient loads would be in a few example catchments under improved land use management and to show how models such as SedNet and ANNEX can be used to help guide management intervention. Three land use scenarios were run using SedNet and ANNEX: (1) A change in ground cover for the Herbert basin under grazing from an average 60 % ground cover to 70 % ground cover; (2) Gully and riverbank revegetation in the Cape River sub-catchment of the Burdekin R. basin; (3) A decrease in fertiliser application in the cane lands of the lower Tully R. basin.

The 70% cover scenario was considered a good estimate of the ‘best practice’ condition for grazing lands in the Herbert basin. The increase in cover resulted in a total sediment export to the estuaries of 563,000 t/y, which is a reduction of ~120,000 t/y of sediment for just a 10 % increase in ground cover. This represents a total decrease in sediment load of 17 % at the mouth of the river. This exercise shows that in the case of the open eucalypt woodland grazing areas, a relatively small change in land use practice can potentially have large impacts on sediment exports as long as it is widely adopted.

The Cape River catchment has an area of 20,000 km2, and makes up 15 % of the Burdekin River basin. The gully revegetation scenario reduced by 16% the predicted amount of sediment delivered from gully erosion across the Burdekin basin, a change from 5100 kt/y to 4300 kt/y. River bank revegetation and gully erosion control reduced the predicted exports by 6%. The reduction in bank erosion is small as much of the catchment currently has reasonably high quality riparian vegetation, according to the input data. The revegetation of riverbanks to 95% of natural extent involves rehabilitating approximately 2300 km of stream. The bank erosion rate is low because of low discharges and low stream gradients. Gully revegetation in the Cape River reduces TN inputs by 400 t/y and TP exports by 100 t/y. The particulate exports at the Burdekin River mouth are reduced by 3%.

The reduction of gully and riverbank erosion in the Cape River is predicted to be not as effective at achieving water quality targets as the Herbert River example for two reasons. First, the change is over a smaller proportion of the total catchment. Second there are other areas of the catchment that contribute more strongly to exports because of higher erosion rates and less potential for deposition during transport.

The main aim of the fertiliser scenario was to assess what happens to the TN exports when the fertiliser application rate is reduced on sugar cane lands. We assessed the result of reducing the current nitrogen application rate of 200 kg/ha/y (made up of fertiliser, mineralization of sugarcane trash nitrogen and a small mill mud addition) to 130 kg/ha/y assuming a resulting 50% decrease in DIN concentrations in surface and subsurface runoff. This scenario results in an 18 % reduction in dissolved N exported from the sugar cane lands and 19 % reduction in dissolved N exports from the basin. The relatively large reduction for management applied to a relatively small area is because of the disproportionately large contribution to N load made by sugar cane land. That is, the management action effectively targets the cause of the problem.

Overall the current model estimates that mean annual export of suspended sediment from the GBRCA is 16 million tonnes, an increase of eight times the estimated natural export of two million tonnes. Nitrogen exports from the GBRCA are 63,000 tonnes, an increase of 4.3 times the estimated natural export of 14,500 tonnes and for phosphorus, the increase is 6.1 times, from a natural export of 1,800 to the current 11,000 tonnes.

Management intervention to reduce nutrient discharge to the GBRWHA in line with the proposed Great Barrier Reef Marine Park Authority (GBRMPA) targets or any revised targets as set under the Reef Protection Plan or regional Natural Resources Management (NRM) plans will be a costly process. Restoration of riparian vegetation, fencing to exclude cattle from streams, wetland re-establishment and/or construction and various forms of fertiliser management will all require substantial funding at the scale of the GBRCA. Effective targeting of available funding will be essential to achieve any effective result. The results of the current modelling, including the scenario modelling predict that the areas of the GBRCA of high priority for management to reduce sediment and nutrient exports are:

1. Areas of high soil erosion, which are predicted to occur along near coastal sub-catchments, are particularly intense along the Mackay-Whitsunday coast, and include the eastern parts of the Fitzroy and Burdekin River basins

2. Areas of moderate delivery of sediment from soil erosion, gully erosion, and occasionally from bank erosion in the Mary, Burnett, Fitzroy, Burdekin, Herbert, and Normanby River
basins, including sub-catchments well inland from the coast.

3. Areas of high soil erosion with soils enriched in phosphorus and nitrogen, which are particularly prevalent along the Mackay Whitsunday coast.

4. Areas of high nutrient loss, particularly dissolved inorganic nitrogen (DIN) loss, on the drainage areas of the larger rivers – Johnstone, Tully, Herbert, Russell-Mulgrave, and lower Burdekin."

---

Citation and/or URL

Brodie, J., McKergow, L.A., Prosser, I.P., Furnas, M., Hughes, A.O. & Hunter, H. 2003, Sources of sediment and nutrient exports to the Great Barrier Reef World Heritage Area, James Cook University, Australian Centre for Tropical Freshwater Research (ACTFR) report: 03/11, Townsville, Australia.

---

Spatial Coverage

Great Barrier Reef-wide 

---

Temporal Coverage

Historical estimations and current (2003) levels 

---

Update Frequency

Not applicable 

---

Other Information

None