Environmental Engineering Reference
In-Depth Information
8.4.1.3 FMC 1989a: corn studies, Nebraska
Farmers under the direction of FMC applied Furadan 4F by air to non-irrigated fi elds at 1.1 kg
ai/ha per application; two applications were made. Although FMC personnel carefully super-
vised the mixing and application, the researchers had considerable diffi culty with their chemical
'accountability'. Tank samples taken directly from the spray booms immediately after spraying
had recovery rates of the active ingredient ranging from 66 to 210%, except for one sample that
contained 1.9% of that expected. The fl owable carbofuran formulation is a suspension that is dif-
fi cult to keep homogenised. The high variability in this carefully monitored fi eld trial suggests that
the amount of carbofuran deposited likely fl uctuates widely under operational use conditions. One
should consider this variability when evaluating residue information from fi eld trials or reported
kills. The researchers also seemed to experience problems with their analytical capability, because
the recoveries of three fresh samples of formulation taken directly from the jugs the product was
sold in ranged from 75 to 84%. Alternatively, this may denote problems with the quality of the
product.
Spray deposit cards placed at various levels in the crop reinforced that there were diffi culties in
chemical accountability. Individual readings from cards placed in the crop canopies ranged from
0.18 to 100% of that expected in the eight treated fi elds over the course of the two applications.
Average deposits from each application ranged from 6.6 to 44%, with a grand mean of 22% of that
expected. Given that a deposit rate of 60% or better is considered typical of a 'successful' aerial
application (Sheehan, Baril, Mineau et al. 1987), this application had poor overall coverage. Peak
residues (two hours post-spray) of fi eld edge vegetation (all four edges being pooled for any given
fi eld, with samples taken 5 metres into the edge) averaged 53.4 ± 64.3 ppm for the fi rst application
and 44.4 ± 90.4 ppm for the second. Field personnel placed drift cards about 3 metres into all four
fi eld edges. The fi eld edge receiving the most drift (presumably the downwind edge, although the
report does not specify wind conditions during application) received between 16.2 and 167% of
mean deposits obtained at the top of the corn canopy (mean of 82.1%, based on 13 different applica-
tions). This extent of edge contamination after aerial application is not surprising; especially where
several spray swathes (i.e., bands) contributed to the downwind spray deposition (Maybank, Yoshida
and Grover 1978).
Field personnel searched the plots for seven days before the fi rst Furadan application and for
seven days after each of the two applications. The study authors used the 'pre-treatment' mortality
as a covariate of post-treatment mortality. However, this was inappropriate because those plots had
been treated with toxic insecticides, including ethyl parathion, between two and three weeks before
the study began. Therefore, the pre-treatment mortality may have been due to pesticide poisoning
rather than refl ecting 'background' mortality. Also, 'pre-treatment' mortality represents the accu-
mulation of carcasses and feather-spots over a prolonged period of time and is a poor covariate for
mortality immediately following a pesticide application. Therefore, it is our view that the statistical
analysis of the carcass data is likely invalid and lacking relevance. The study authors also compared
the kill data with surveys of live birds conducted on the plots. Such comparison provides little to no
understanding of the pesticide effect, and the authors stated that the live bird surveys were diffi cult
to interpret because the lateness of the season meant that the bird populations were mobile and not
'tied' to the study fi elds.
However, as the areas searched in treated and control fi elds were similar and search intensities
were held roughly constant, it is still meaningful to look at the overall number of carcasses found in
treated versus control fi elds, assuming that fi eld personnel found most of the pre-treatment carcasses
over the seven days of searching. This effectively assumes that the personnel had swept clean the
areas of carcasses before the application of Furadan 4F. For reasons given above, the number of car-
casses found during these pre-treatment sweeps to assess natural mortality is not relevant. In other
words, this study is potentially useful from a qualitative standpoint only.
Search WWH ::
Custom Search