Article pubs.acs.org/JAFC
Spot Spraying Reduces Herbicide Concentrations in Runoff Alice R. Melland,†,* D. Mark Silburn,†,§ Allen D. McHugh,# Emilie Fillols,⊥ Samuel Rojas-Ponce,§ Craig Baillie,† and Stephen Lewis× †
National Centre for Engineering in Agriculture, University of Southern Queensland, Toowoomba, 4350 Queensland, Australia Department of Natural Resources and Mines, Queensland Government, Toowoomba, 4350 Queensland, Australia # International Maize and Wheat Improvement Centre (CIMMYT), Ningxia Academy of Agriculture and Forestry Sciences, Yinchuan, Ningxia, People’s Republic of China ⊥ Sugar Research Australia Ltd., Te Kowai, 4740 Queensland, Australia × Catchment to Reef Research Group, TropWATER, James Cook University, Townsville, 4811 Queensland, Australia §
S Supporting Information *
ABSTRACT: Rainfall simulator trials were conducted on sugar cane paddocks across dry-tropical and subtropical Queensland, Australia, to examine the potential for spot spraying to reduce herbicide losses in runoff. Recommended rates of the herbicides glyphosate, 2,4-D, fluoroxypyr, atrazine, and diuron were sprayed onto 0, 20, 40, 50, 70, or 100% of the area of runoff plots. Simulated rainfall was applied 2 days after spraying to induce runoff at one plant cane and three ratoon crop sites. Over 50% of all herbicides were transported in the dissolved phase of runoff, regardless of the herbicide’s sediment−water partition coefficient. For most sites and herbicides, runoff herbicide concentrations decreased with decreasing spray coverage and with decreasing herbicide load in the soil and cane residues. Importantly, sites with higher infiltration prior to runoff and lower total runoff had lower runoff herbicide concentrations. KEYWORDS: pesticide, water quality, sugar cane, precision spraying, postemergent, pre-emergent, sediment−water partition coefficient
■
INTRODUCTION Pesticide runoff from tropical agriculture can adversely affect receiving water bodies1−3 and is a significant threat to water quality in the Great Barrier Reef (GBR) World Heritage Area of tropical northeast Australia.4 Although concentrations of pesticides in waters around the reefs are generally low,5,6 guideline concentrations7 are often exceeded by 10−50 times in catchment rivers.8 The sugar cane industry is one of the largest users of pesticides in the GBR catchments and, despite occupying only 1.3% of the GBR catchment area, the industry has been identified through modeling (94% of total loads,9) and measurement10−12 as the most significant contributor of the commonly detected photosystem II-inhibiting (PSII) herbicides to the GBR. To mitigate detrimental impacts of pesticides on the health and resilience of the GBR, targets for water quality and land management have been set.13,14 The current water quality target for pesticides is for at least a 60% reduction, compared to the baseline year of 2008−2009, in end-of-catchment pesticide loads in priority regions by 2018. The average annual pesticide load was modeled to have reduced by 28% (4.6 t year−1) from 2008 to 2013, with 80% of the reduction occurring in the Wet Tropics and Mackay Whitsunday sugar cane regions due to improved pesticide management.9 Between 2009 and 2011, land management practices were improved by 34% of sugar cane growers, which demonstrates progress toward the target of 90% of priority land being managed using best management practice systems.14 Weed control regimens in the Australian sugar cane industry include a combination of herbicide application, mechanical © XXXX American Chemical Society
cultivation, and retention of a green cane trash (cane residue) blanket (GCTB) on the soil surface after harvest.15 Herbicides that are currently widely used in the industry include the PSII herbicides atrazine, diuron, and ametryn for pre-emergent “residual” weed control (acting via soil exposure) and “knockdown” herbicides 2,4-D, fluoroxypyr, and glyphosate for postemergent control (acting via contact with plants).16 Increased adoption of minimum tillage since the 1980s reduced the use of mechanical cultivation for weed control in plant cane, but increased the reliance on herbicides, particularly atrazine.17 Improved practices that have been identified from research as “best management” for minimizing off-site herbicide transport in runoff from cane farms include a range of soil, crop, water, and herbicide management practices.18 The water quality benefits of many of these practices including GCTB,19 controlled traffic farming,20−22 and timing herbicide application to avoid forecast heavy rain21 are supported by studies in the GBR catchments23 and elsewhere. To add to this suite of beneficial practices, practices that reduce the total volume of herbicide applied to paddocks are also advocated18 and were the focus of this research. Traditionally, herbicides in sugar cane have been broadcast sprayed to both the planted row or bed and the inter-row or furrow, resulting in 100% (or blanket) coverage of the paddock. Special Issue: Pesticide Fate and Effects in the Tropics Received: July 28, 2015 Revised: October 14, 2015 Accepted: October 19, 2015
A
DOI: 10.1021/acs.jafc.5b03688 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
B
20 1.40
Australian Soil Classification.32 bSurface soil (0−10 cm) properties. cWeed cover broadly correlated with 20, 50, and 70% spray coverages. a
0, 40, 50, 100 0, 20, 50, 100
glyphosate, 2,4-D, fluoroxypyr atrazine diuron 0, 20, 50, 70, 100 Mkay_2012
Mackay-Whitsunday, Mackay, subtropics, rain-fed
Sodosol
plant
bare soil and weeds
0, 20, 50, 100 0, 40, 50, 100
0, 20, 50, 70, 100 bare soil and trash ratoon Brown Dermosol Burdekin, Mulgrave, dry tropics, furrow irrigated Bkin_2012
0, 20, 50, 100
glyphosate, 2,4-D, fluoroxypyr diuron atrazine
sandy light medium clay
sandy clay loam
1.40
1.39
32 1.66
18 1.94 1.50 Burnett-Mary, Bundaberg, subtropics, rain-fed Bberg_2012
Brown Dermosol
ratoon
trash
0, 20, 50, 100
glyphosate, 2,4-D, fluoroxypyr atrazine, diuron
fine sandy clay loam
1.63 1.56 0, 20, 50, 70, 100 trash and weedsc ratoon Brown Sodosol Mackay-Whitsunday, Mackay, subtropics, rain-fed Mkay_2011
soil typea region, district, climate, and water supply system
glyphosate, 2,4-D, diuron
fine sandy loam
soil organic Cb (%) soil bulk densityb (g cm−3) soil textureb herbicides applied herbicide spray coverage rates (%)
Sites, Soils, and Field Conditions. Plot-scale runoff trials were conducted at five sites across the intensive sugar cane growing regions on the coastal plain of Queensland, Australia (Table 1). The sites were selected to represent a range of management systems, climates, and
crop phase and water source
ground cover type
MATERIALS AND METHODS
site_year
Table 1. Site, Crop, Soil, and Herbicide Treatment Characteristics
■
clayb (%)
However, precision agriculture technologies now available enable more spatially targeted spraying which, importantly, can reduce the volume of herbicide required for weed control. For sugar cane, applying residual herbicides in bands to the row (and not the inter-row) and using shields to protect an established cane crop from knockdown herbicides applied to inter-rows reduced the total herbicide volume applied to a paddock by 40−60% compared with broadcast application.21,24 A further advancement to shielded band spraying is automated spot-spray technology, which uses machine vision to discriminate weeds from bare soil (e.g., Weedseeker), and more recently in research trials, to discriminate weeds from crops.25 Spot spraying can reduce the applied volume of knockdown herbicides by ≥80% compared with broadcast spraying, because the area requiring spray is dependent only on the weed pressure. Experimental work in sugar cane suggests spot spraying can also reduce the cost of weed control relative to conventional techniques,26 although the cost of the technology currently limits widespread commercial adoption. Precision spraying technologies also increase the feasibility for nonselective knockdown herbicides such as glyphosate to be used in established crops with minimum harm to the crop. From an environmental perspective, knockdowns are preferred to residual herbicides because they often have shorter half-lives for degradation in soil, are more readily adsorbed to soil, and have less and shorter-lived off-site environmental impact.27,28 The knockdown herbicide glyphosate, for example, is about 2000 times less potent than the herbicide diuron, which has both knockdown and residual activities.29 However, knockdown herbicides may become problematic in non-target aquatic environments if used in large quantities,30 and overly frequent use can promote herbicide resistance in weeds.31 Using precision spraying to reduce the coverage and therefore volume of all herbicides applied, without reducing application rates and weed control efficacy, is also critical. The effectiveness of reducing the area of the paddock sprayed in reducing herbicide runoff losses has been tested for a limited range of tropical soils, herbicides, and row crops. In sugar cane, Masters et al.21 found that banded spraying of residual herbicides onto the row beds of a ratoon crop (i.e., 50−60% of the paddock) with GCTB cover reduced rainfall runoff herbicide concentrations proportionately (48−57%) compared with blanket spraying. Even larger reductions occurred in irrigation runoff from sugar cane furrows, with application of residual herbicides to row-beds only (i.e., 40% of the paddock), resulting in a 90% reduction in runoff herbicide loads compared with blanket application.24 The effectiveness of spot-spray technology on herbicide loss in runoff, however, has not been reported. The objective of this study was to evaluate the potential for spot-spray technology to reduce the loss of herbicides in runoff compared with banded and blanket spraying across a range of sugar cane soils and locations. To meet this objective, a rainfall simulator was used to generate runoff from small plots after herbicides suited to precision spray methods were applied to various proportions of the plot areas in ratoon (GCTB and bare soil) and plant (bare soil) cane fields.
40 mm h−1) natural rainfall and are mounted on an oscillating manifold. Nozzle pressure was regulated to 60 kPa, and the rainfall intensity was controlled by regulating the frequency of successive spray passes across the plots. Rain was applied at rates (70−80 mm h−1, Table 2) representing a 30 min rainfall intensity that had a 1 in 2 year recurrence interval. Cane plants between the plots were mowed to 0.05). bResponse variate loge transformed.
Figure 4. Relationship between site runoff characteristics (volume of rainfall that infiltrated before runoff started and total percentage of applied rainfall that infiltrated) and the propensity for herbicide loss, represented by the slopes of significant (P < 0.05) linear relationships between dissolved herbicide EMC and the soil and trash herbicide loads. Slopes were derived from comparisons between sites for each herbicide using data that were normalized for the maximum herbicide EMC, and soil and trash loads, across sites.
influence of site and soil hydrology highlights the importance of
into the soil, and degradation to minimize mobilization in runoff.21 Processes of Herbicide Mobilization. The majority (>50%) of all herbicides in runoff from the two sites with
increasing the time between herbicide application and irrigation or forecast rain to allow for herbicide uptake by weeds, leaching H
DOI: 10.1021/acs.jafc.5b03688 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry
Table 5. Pesticide Leaching and Partitioning Properties either Reported in the Literature or Measured in This Experiment
pesticide leaching potential (PLP)b
active ingredient glyphosate AMPA 2,4-D (amine) fluoroxypyr atrazine diuron
pesticide solution runoff potential (PSRP)b
pesticide adsorbed runoff potential (PARP)b
toxicity relative to diuronc
soil sorption Kdd (L kg−1)
low
high
high
0.00052
325
intermediate intermediate high intermediate
high high high high
low low intermediate intermediate
0.011 0.0036 0.15 1
1 3 2 16
sediment−water partition coefficient Kpa (L kg−1)
percentage herbicide EMC in dissolved phase, mean of all plots where sediment herbicide content was measured (% ± SE)
Mkay_2012
Bkin_2012
Mkay_2012
187 240 0.4 4 3 17
71 219 21 29 12 110
60 53 100 97 98 89
± ± ± ± ± ±
11.8 10.4 0.1 1.5 0.3 2.6
Bkin_2012 82 68 93 91 97 64
± ± ± ± ± ±
2.0 5.5 0.9 1.4 0.3 4.9
Calculated from concentrations of herbicide in the filtered (
