Minimising runoff and sediment loss . Given that pesticides are transported in run-off ( in their dissolved phase ) or attached to sediments ( their sorbed phase ), reducing run-off and sediment losses can contribute to managing pesticide exports from fields . Practices that reduce run-off and sediment movement , such as retention of crop residues and controlled traffic , are well established ( Freebairn et al ., 1996 ) and their effectiveness in reducing pesticide losses have been confirmed in studies in the Lower Burdekin ( Cowie et al ., 2012 , 2013 ; Silburn et al ., 2013a ).
On irrigated farms , off-site herbicide losses are greatest in the first few irrigation events and taper off as irrigations continue throughout the crop growth period ( Davis et al ., 2014 ). Tail water capture and recycling provide useful control of paddock run-off during dry times ( Davis , 2013 ; DeBose et al ., 2014 ). Tail water recycling is widely practised on a significant proportion of sugarcane farms in the BRIA ( Shannon and McShane , 2013 ). See Section 4.2.3 for further information .
Improvements to the paddock scale modelling that have been incorporated in recent years include four regionalised scenarios that encompass a suite of traditional PSII herbicides and the alternative herbicides that are perceived to be less toxic to corals and seagrasses . The rate of applications is the dominant contributing factor , with limited differences associated with different irrigation techniques . Considering that the spatial risk of herbicide loss is largely homogenous , it is recommended that efforts be directed into the rate of PSII herbicide application , which can be improved by uptake of banded application methods across both BRIA and Delta regions .
Choosing products with rapid degradation rates ( e . g . some ‘ knockdown ’ herbicides ). Choice of product is a complicated area of herbicide management because to achieve improved water quality , the alternative product needs to be equally or more effective in weed control , be no more prone to run-off losses and / or be no more toxic than the current product ( Davis et al ., 2014 ). All the above requirements are currently subject to uncertainties . In recent years there has been a shift within the sugarcane industry towards the use of alternatives herbicides such as metolachlor , impazipic , isoxaflutole , metrabuzin and pendimethalin . These are perceived to have a reduced toxicity , mobility and persistence compared with PSII herbicides ( Davis et al . 2014 ). While relatively little is known about the long term acute and chronic effects when organisms in freshwater and marine environments are exposed to these alternative herbicides ( Davis et al . 2014 ), major progress has been made in understanding the relative run-off risk and relative toxicity of older and emerging herbicides , but these are still works in progress ( Lewis et al ., 2013 , 2016 ; Devlin et al ., 2015 ; Smith et al ., 2016a , 2016b ).
Most of the older , soil residual ( e . g . PSII and other ) herbicides have somewhat similar physicochemical properties and application rates and similar high susceptibility to loss in run-off ( e . g . Cowie et al ., 2012 ; Melland et al ., 2016 ; Silburn et al ., 2013b , 2013c ). The products that are less prone to losses through run-off include those applied at lower rates ( e . g . imazapic , isoxaflutole and fluroxypyr ), those with greater soil sorption ( e . g . paraquat , pendimethalin ) and those with shorter half-lives ( e . g . glyphosate and 2,4-D ), as illustrated by modelling using generic properties ( Shaw et al ., 2011 ). Experience in grain and oilseed cropping in the United States has also found that replacing soil residual herbicides with knockdown herbicides ( e . g . in glyphosate-resistant crops ) resulted in less herbicide run-off and less toxic loads , in both monitoring ( Shipitalo et al ., 2008 ) and modelling ( Wauchope et al ., 2002 ).
A series of rainfall simulation trials ( Cowie et al ., 2012 ; Melland et al ., 2016 ; Silburn et al ., 2013b , 2013c ) on a variety of herbicides ( older residuals , alternative residuals and knockdowns ) at two days after application ( before half-life differences emerge ) showed that run-off loads of the older and alternative residual herbicides and 2,4-D were reasonably similar and high . Herbicides applied at much lower rates ( e . g . fluroxypyr , imazapic , isoxaflutole ) had lower run-off loads . Glyphosate run-off loads were generally lower , and for pendimethalin were considerably lower , than for residuals applied at similar rates . Results varied depending on site conditions ( e . g . soil type and cover ) and hydrology . This is still a work in progress . An overarching framework for explaining the differences quantitatively is yet to be found , particularly for soils or field conditions with contrasting hydrology .
25