The results also demonstrated that glyphosate sprayed on the soil is still reasonably prone to run-off even though its soil sorption is greater than many soil residual herbicides ( Devlin et al ., 2015 ). Glyphosate has been a cornerstone of industry transitions towards minimum tillage agriculture and lower dependence on residual PSII herbicides . However , it is being detected frequently at catchment scales since its recent addition to Great Barrier Reef monitoring programs ( Wallace et al ., 2016 ), and greater knowledge of its environmental behaviour is needed . Weed resistance to glyphosate ( and to other herbicides ) is also driving large changes in weed control practices and will continue to do so in future .
Lewis et al . ( 2013 ) have attempted to develop a risk assessment framework based on ( i ) run-off loads of herbicides under simulated rainfall ( Silburn et al ., 2013b , 2013c ; Melland et al ., 2016 ), ( ii ) herbicide dissipation rates in soil ( Shaw et al ., 2013 ) and ( iii ) the relative toxicity of herbicides to aquatic plants ( Poggio et al ., 2014 ; Smith et al ., 2016a , 2016b ). When toxicity and run-off potential are combined , most herbicides had less risk than diuron , although ametryn had a higher risk . The reduced risks relative to diuron are reinforced if rainfall occurred 30 days after application rather than at two days . While this analysis is preliminary and requires completion of current studies on toxicity and further direct comparisons of run-off potential and half-lives for more products , it does illustrate a way forward in providing useful advice on product choice .
4.2.3 Increasing irrigation efficiency and reducing runoff
Increasing irrigation efficiency ( i . e . reducing over-application of irrigation ) reduces nutrient and pesticide losses and is a critical action for the Lower Burdekin catchment . Improvements in water use efficiency are predicted to result in reductions in nitrogen loss especially in the dry season , and particularly in the heavier BRIA soils ( Thorburn et al ., 2011b and confirmed in recent data supplied by the Paddock to Reef program ). Excess irrigation water also has implications for the volume of surface runoff which is greater if the soil profile is ‘ full ’. Therefore , irrigation efficiencies may be considered more important in the BRIA where there is comparably less deep drainage than in the Delta and greater capacity for surface runoff ( Thorburn et al ., 2011b ).
Irrigation practices are defined in the P2R Water Quality Risk Framework in terms of : 1 ) calculating the timing and amount of water to supply , 2 ) optimising the irrigation system , 3 ) managing surface runoff . Techniques for addressing these practices are outlined in the following sections .
Matching irrigation applications as closely as possible to crop water requirements . Within the P2R Water Quality Risk Framework , calculating the amount of water to supply is the most important management tactic for water quality outcomes with best management practice requiring that the amount of irrigation water applied to each block is less than or matches the soil water deficit . In a simulation study of furrow-irrigated sugarcane in the Lower Burdekin , reducing total application of irrigation water from approximately 4,000 to approximately 2,000 mm / yr was predicted to reduce nitrogen losses by run-off or deep drainage by approximately 5 – 50 per cent ( depending on the soil type ) without reducing crop yields ( Thorburn et al ., 2011b ). Further reductions in irrigation applications further reduced losses but also reduced yields .
It is established that 95 per cent of the sugarcane area in the Lower Burdekin is irrigated using furrow irrigation systems ( see Section 6.1.1 ), there is a lack of data on how these irrigation systems are operating in terms of application efficiency and total water usage versus losses . However , long run average crop irrigation requirements are generally considered to be at least 8 ML / ha using efficient furrow irrigation ( including laser levelled paddocks etc ). Data on the range of irrigation application efficiencies ( i . e . the proportion of the water applied that is used by the crop and not lost to runoff or drainage ) is important for understanding water quality influences from irrigation practices and is currently limited in the GBR . The Paddock to Reef program defines a set of application efficiencies for different practice levels in the BRIA and the Delta , shown in Table 4.2 . This indicates that the application efficiencies likely to be achieved in the BRIA are likely to be higher than those in the Delta .
26