3.2.1 Dissolved inorganic nitrogen
The estimated annual average anthropogenic load of DIN from the Lower Burdekin sugarcane area using the Source Catchments model ( 2013 baseline ) is 1,046 tonnes per year , or 87 per cent of the regional anthropogenic load . These estimates correlate well with recent monitoring data ( Turner et al ., 2013 ; Garzon- Garcia et al ., 2015 ). Nitrogen losses in surface runoff and deep drainage vary according to soil type in both BRIA and Delta regions . The dominant loss pathway in the largely clay based soils of the BRIA is through surface water runoff whereas in the highly permeable light soils of the Delta the dominant loss pathway is through deep drainage ( Thorburn et al ., 2011b ).
The BRIA and Delta regions contribute approximately equal loads of DIN ( accounting for uncertainties in some of the model input data such as current management adoption ). In 2016 , this was estimated at 460 tonnes / year ( 44 per cent ) and 586 tonnes / year ( 56 per cent ) respectively , with both areas occupying approximately equivalent proportions of the cane producing area ( CPA ) in the Lower Burdekin catchment . However , the DIN export rates are modelled as being higher in the Delta (~ 14 kg / ha / yr ) compared to the BRIA (~ 10 kg / ha / yr ). Specific modelling results at this scale are currently being updated and were not available at the time of preparing this report ( see also Section 5 ).
3.2.2 PSII herbicides
Sugarcane is the greatest contributor of PSII herbicide exported load in the Burdekin region , contributing a majority ( 99 per cent ) of the regional load which is estimated to be 2,295 kg per year ( McCloskey et al . 2017 ). This equates to a Diuron Toxic Equivalent load of approximately 2,100 kg per year . The modelled PSII herbicide toxic equivalent load data for the region also shows that the modelled contributions from the BRIA and Delta are roughly equivalent ( within the level of accuracy of the model ), estimated at 693 kg / year ( 55 per cent ) and 564 kg / year ( 45 per cent ) respectively .
The lack of distinction between regions is likely a result of the reporting of relatively homogenous farming practices across both BRIA and Delta , and the factors that influence pesticides loss processes do not vary significantly across regions . However , it is likely that more residuals are applied in the BRIA as standard practices use less cultivation and therefore weed pressure can be greater ( E . Shannon , pers . comm .). In addition , the loss pathways are different , and a large proportion of the irrigation tailwater runoff in the BRIA discharges to Barratta Creek , compared to the Delta where the waterways are typically used as water transfer channels receiving surface and groundwater flows ( Davis et al ., 2012 ).
The 2017 Scientific Consensus Statement included a risk assessment of the PSII herbicides at the end-ofsystem monitored sites in region ( see Waterhouse et al ., 2017b ). Monitoring data for Barratta Creek between 2010 and 2016 had a Very High risk score in 2013-2014 and High scores in all other years except 2010-2011 which was Moderate ( Table 3.2 ). For the freshwater and estuarine ecosystems , a score of Good is required for the protection of these ecosystems as they are considered to be slightly to moderately disturbed ecosystems , but for the Great Barrier Reef World Heritage Area a score of Very Good is required as it is a high ecological value system . Thus , these results indicate a risk from pesticides in Barratta Creek catchment and the downstream marine ecosystems are potentially at risk from pesticides , based on the point in the catchment that was monitored . It should be acknowledged that only a proportion of all catchments have been assessed here due to the limitations of the monitoring data .
An assessment of exceedances of ecotoxicity threshold values for 21 pesticides detected in monitoring between 2013 and 2016 was also completed ( see Waterhouse et al ., 2017b ). Similar to the assessment , Very High scores occurred in Barratta Creek for diuron ( all years ) and imidacloprid ( 2014 and 2015 ), with High and Moderate scores for metolachlor in 2014 and 2015 respectively . Diuron and metolachlor are PSII herbicides used for weed management and imidachloprid is highly toxic insecticide used to treat cane grub . A High score also occurred for diuron in the Haughton River in 2015 ( Table 3.3 ). The high total number of pesticides detected in Barratta Creek and the Haughton River ( shown by the coloured boxes in Table 3.3 ) is important as the small effect from each pesticide is cumulative , resulting in a higher overall risk from the mixture as a whole .
12