We tracked dicamba vapor drift by conducting experiments using field soybeans and greenhouse-grown bioassay soybeans. Research was conducted in 2009 and 2010 at the Russell Larson Agricultural Experiment Station in Rock Springs, Pennsylvania, USA. In each year, six (2009) or eight (2010) fields were planted with a dicamba-susceptible, glyphosate-resistant cultivar (Chemgrow #3539) on staggered dates between mid-May and early June. Soybeans were no-till drilled in 19-cm rows in fields that had been planted with corn the previous year. Weed control was provided with a preplant application of chloransulam-methyl at 18 g active ingredient/ha and an early postemergence application of glyphosate at 1,262 g acid equivalent/ha. We then applied dicamba treatments to each field in late June to mid July, while soybean were in the V3 to V5 stage 17, which is a typical timing for postemergence herbicide applications on soybean. These treatment dates covered a range of environmental conditions that may influence vapor drift, including air pressure, relative humidity, air temperature, and wind speed. Therefore, the results span the likely range of vapor drift from postemergence applications on soybean.
In the center of each field, we treated an 18.3 × 18.3 m plot with a dicamba herbicide at a rate of 561 g acid equivalent/ha, applied with a total carrier volume of 187 L/ha. We used 561 g acid equivalent/ha because this is currently a common application rate in corn and is also within the range of rates that will be promoted for postemergence applications in dicamba resistant soybean 8, 9, 18. In 2009, we treated all six fields with a formulated product containing the dimethylamine salt of dicamba 8. In 2010, we treated five fields with dimethylamine dicamba and three fields with a formulated product containing the diglycolamine salt of dicamba 18. We applied herbicides using a backpack CO2 sprayer and a 3.05 m handheld boom equipped with AIXR11002 nozzles (TeeJet), calibrated to 241 kPa, and walked on at a speed of 1.3 m/s. Herbicide was applied to each field between 7:45 and 8:30 AM.
In advance of the herbicide application, we extended eight transects from the center of the treated plot in each field. These transects stretched from 25 to 90 m, depending on field size and shape. Designating the center of the treated plot as the zero point, we marked individual field soybean plants for recording dicamba herbicide injury symptoms every 3 m up to 25 m, every 5 m up to 40 m, and then every 10 m up to 100 m. To gauge variability at a single location, we marked three plants at 6, 12, and 21 m along each transect. We quantified injury symptoms on marked plants 28 d after herbicide treatment on a 100-point scale using a scale similar to that developed by Behrens and Lueschen 4 and Anderson et al. 19 (Table 1).
Table 1. Dicamba injury symptom rating scalea
|0||No effect, plant normal.|
|10||Slight crinkle of leaflets of terminal leaf.|
|20||Cupping of terminal leaflets, slight crinkle of leaflets of second leaf, growth rate normal.|
|30||Leaflets of two terminal leaves cupped, expansion of terminal leaf suppressed slightly.|
|40||Malformation and growth suppression of two terminal leaves, terminal leaf size less than one-half that of control. New axillary leaves developing at a substantially reduced rate.|
|50||No expansion of terminal leaf, second leaf size one-half that of control. Axillary leaf buds unable to open and develop.|
|60||Slight terminal growth, necrosis of terminal leaf, and axillary bud apparent. Chlorosis and necrosis in axillary leaf clusters.|
|70||Terminal bud dead, substantial, strongly malformed axillary shoot growth.|
|80||Limited axillary shoot growth, leaves present at time of treatment chlorotic with slight necrosis.|
|90||Plant dying, leaves mostly necrotic.|
Injury on field soybean plants integrates both particle drift immediately following herbicide application and vapor drift potentially occurring over a longer period. To exclude particle drift and more precisely measure vapor drift occurring in the period following application, we used potted soybeans as bioassay plants. We grew sets of bioassay plants in a greenhouse under ambient light conditions, with one set for each treated field seeded one to three weeks following the planting of the corresponding field soybean. Greenhouse bioassay plants were grown in 3.8-L pots using an artificial soil mixture (Fafard pro germination mix) supplemented with 4.0 g of Osmocote 19-6-12 fertilizer (Scotts Miracle-Gro Company). The same soybean cultivar was planted in the field and greenhouse. Depending on field size, we grew between 104 and 144 bioassay plants per set. We randomly assigned each plant a position within a grid along a greenhouse table, with plants designated for the same field grouped on the same tables. We also randomly assigned each plant a position on one of the transects in its corresponding field. Greenhouse bioassay plants were assigned to the same positions as the field plants, with three plants also assigned to the 6, 12, and 21 m positions along each transect. We scored the bioassay plants for growth stage and measured height to the apical bud on the day immediately preceding field exposure.
On the morning of each dicamba field application, we transported the bioassay plants from the greenhouse to the designated field. We parked the vehicle more than 100 m away from the treated plot, with all doors and windows closed during application. Thirty minutes after herbicide application (allowing time for particle drift to settle), a crew of four to six research technicians positioned the bioassay plants within approximately 20 cm of the marked field plants along the eight transects. In 2009, we collected the bioassay plants after 24 h of exposure in the field and then transported them back to the greenhouse. In 2010, complications with field operations (overnight rodent and deer damage) forced us to switch from a 24-h to 8-h exposure midway through the experiment. We then scored each bioassay plant for dicamba injury symptoms 14 d after herbicide treatment using the same scale used for field soybeans. A summary of the date of treatment, formulation, and exposure duration of bioassay plants for each field is provided in Table 2.
Table 2. Summary of field trials by formulation, duration of exposure, treatment date, number of bioassay plants, and number of false positive recordings
|Field||Formulation||Exposure (h)||Treatment date||No. Plants||False-positives|
|KN4W||DMA||24||June 27, 2009||104||3|
|H||DMA||24||July 1, 2009||132||5|
|KN5||DMA||24||July 6, 2009||120||2|
|Ento||DMA||24||July 8, 2009||120||2|
|FryJ1||DMA||24||July 10, 2009||120||6|
|KN4E||DMA||24||July 16, 2009||120||1|
|KN6W||DMA||24||June 17, 2010||144||2|
|U8||DMA||24||June 23, 2010||112||1|
|KNE2||DMA||8||June 26, 2010||140||4|
|OMN||DMA||8||July 6, 2010||144||2|
|HarpAE||DMA||8||July 21, 2010||128||7|
|KN6E||DGA||24||June 18, 2010||144||2|
|G24||DGA||8||June 25, 2010||136||2|
|HarpAW||DGA||8||July 14, 2010||125||5|
Several environmental variables were collected from each field trial to be included as potential explanatory variables in analysis of the data. A National Oceanic and Atmospheric Administration weather station at the research farm 20 located 100 to 2,100 m from each field collected temperature, air pressure, humidity, precipitation, and wind speed data in 1-min increments.
To account for possible stresses of moving bioassay plants into the field and for possible inadvertent dicamba contamination, we included at least four plants each of two types of controls in each set of greenhouse bioassay plants. “Moved” controls were transported out to the research farm along with the other bioassay plants, but positioned in an untreated field at least 1,000 m from the dicamba application. “Moved” plants were left in the field for the same period as the corresponding bioassay plants and transported back to the greenhouse in the same van. Plants that were “not moved” were left in the greenhouse for the duration of the experiment. We scored both types of controls for injury at the same time intervals as the other bioassay plants in a set. Very slight injury symptoms were observed on one “not moved” plant in 2009 and one “not moved” plant in 2010 (ratings of 5 and 8 on our scale, respectively), and we suspected these observations as being false-positives.
Controlled dose–response experiments
We conducted dose–response experiments to relate injury symptoms observed on greenhouse bioassay plants to the amount of dicamba exposure. In 2009, a randomized complete block design experiment was conducted using six rates (56.1, 5.61, 0.561, 0.056, 0.006 g acid equivalent/ha plus a water control) of dimethylamine dicamba and 12 replications. Seeds were planted on September 29, 2009, using the same methods for the plants positioned in the field. Dicamba was applied in a closed chamber using a track sprayer on October 29, 2009, while plants were in the V3 stage. In 2010, a similar experiment was conducted using 10 rates (561, 56.1, 17.7, 5.61, 1.77, 0.561, 0.177, 0.056, 0.018, 0.006 g acid equivalent/ha plus a water control), two formulations (dimethylamine and diglycolamine), and five replications. Plants were seeded on June 9 and treated on July 2 using the track sprayer, while plants were in the V4 stage. In both years, we quantified injury symptoms on all plants 14 d after herbicide treatment.