Modelling, monitoring and sediment tracing in the Tully river catchment, North Queensland: a comparison of techniques

The Great Barrier Reef (GBR) lagoon is an internationally significant ecosystem, located off the Queensland coast. The Reef Water Quality Protection Plan aims to halt and reverse the decline in water quality entering the reef. One of its actions requires the development of Water Quality Improvement Plans (WQIP). As sediment is one the key pollutants identified within the Tully catchment, its sources and the impact of changed management need to be identified as part of the WQIP. A range of techniques are available to ascertain the various sources and volumes of sediment being delivered and transported in a catchment. Three techniques include modelling, water quality monitoring and sediment tracing. Each has its advantages and disadvantages but when carried out simultaneously can provide valuable information. We compare and contrast these techniques as part of the WQIP research component. Catchment scale models such as the Sediment and River Network Model (SedNet) can provide an understanding of how catchments function by constructing a sediment budget. The three studies showed similar results relating to sediment erosion and transport. The total modelled sediment load supplied to streams was 183 kt/yr. Hillslope erosion was the dominant process with 116 kt/yr or 64% of the total sediment supply. Contributions from bank and gully erosion were 57 kt/yr (31%) and 10 kt/yr (5%) respectively. Monitored Total Suspended Solid (TSS) concentrations from all sub-catchments were generally low; most < 25 mg/L. Forest sites were consistently lower than disturbed sub-catchments such as sugarcane and bananas. A comparison of mean monitored TSS concentrations and modelled annual mean concentrations from SedNet showed reasonable agreement. Results from sediment tracing showed that hillslope erosion was dominant contributor to suspended sediment (75%) comparing well with modelled and monitored data. In terms of landuse, the results from the three studies generally show that a high proportion of sediment is coming from sugarcane. However, the SedNet model over estimated the sediment load derived from forest sources. It is acknowledged that the Revised Universal Soil Loss Equation is not particularly suitable for the prediction of hillslope erosion from steep naturally forested areas under wet tropical conditions. These techniques assist to build robust knowledge about catchment behaviour. Modelling has the major advantage that it is fairly easy, cost effective and has the ability to run ‘what if’ scenarios that the other techniques cannot. Total suspended sediment concentrations in the Tully River are low in comparison with other north Queensland rivers despite a high rainfall. This is probably because most of the Tully WQIP region is relatively undisturbed forest. Of the suspended sediment being transported by the river, most originated from agricultural landuse. This comparison of studies corresponds well with other erosion and water quality studies in the Wet Tropics. Understanding sediment sources will assist with the targeting of limited resources towards reducing soil erosion