Impacts of carbofuran on birds in Canada and the United State - Carbofuran and Wildlife Poisoning: Global Perspectives and Forensic Approaches - page 229

actually received the same insecticide application. Furthermore, there were insecticide treatments
(including carbofuran use) in some fi elds next to study plot fi elds, again confounding the exposure
situation. Nevertheless, we provide highlights of these studies.
Due to the lag between the two treatments, there were two periods of pre-treatment carcass
searches in both the Kansas and Oklahoma fi elds. Field personnel carried out searches for ten days
before each treatment and for seven days thereafter.
Samples taken directly from the tank sprayer showed carbofuran recoveries ranging from 46 to
106% (N 8), again suggesting poor mixing of the active ingredient in the spray solution. Residues
collected on cards just above the crop canopy showed that deposits on individual plots ranged
between 23 and 57% of that expected after the fi rst application and between 1.0 and 75% of expected
for the second. Grand means were 43 and 23% for the fi rst and second sprays, respectively. These
deposits are low for ground application equipment. The highest fi eld edge deposits (presumably the
downwind samples, although their location was not specifi ed) taken three metres in from the fi eld
edge ranged from barely detectable to 14 times higher than those obtained in the fi elds themselves.
In six of the 16 applications monitored, maximum fi eld edge residues were higher than the average
fi eld deposit.
Field personnel found only fi ve bird carcasses over the 20 days of pre-treatment periods in all
fi elds, whether treated or control. However, personnel found 15 dead birds in the search areas after
the two carbofuran sprays and seven after the two 'control' chlorpyrifos sprays. Personnel collected
eight of the 15 birds found on carbofuran plots post-treatment after the second spray at the lower
application rate. Searchers also found four immobilised individuals, three in carbofuran-treated
plots. These included a northern harrier that was seen feeding on an eastern cottontail (a small
rabbit, Sylvilagus fl oridanus ) which contained 0.1 ppm of carbofuran residues. Field personnel also
found non-bird carcasses in both carbofuran- and chlorpyrifos-treated plots, but only after insecti-
cide application. Of the eight bird carcasses found in either post-treatment period on the carbofuran
plots and analysed, fi ve had detectable carbofuran in the gastrointestinal tract or brain. This is very
signifi cant in light of the aforementioned diffi culty of recovering carbofuran from spiked carcasses
(refer back to Section 8.4.1.3). Likewise, two of three carcasses found in either post-treatment period
in the control fi elds contained chlorpyrifos residues. The signifi cance of this fi nding is more diffi cult
to ascertain, because the study did not include fi eld spiking carcasses with chlorpyrifos.
It appears that both the carbofuran and chlorpyrifos treatments caused mortality in this study. The
uncorrected kill rate for carbofuran was 0.47 birds per hectare.
8.4.1.6 FMC 1989b, Oklahoma fi elds, aerial application
This study reaffi rmed the diffi culty of controlling carbofuran applications. The recovery rate of sam-
ples taken from the spray tanks after application ranged from 23 to 189% of that expected (N 6).
Readings from deposition cards in the fi elds ranged from 8.5 to 56% of the nominal rate during the
fi rst application and from 1.2 to 47% of nominal during the second. As in the Kansas ground spray-
ing situation, deposits were extremely low in some fi elds, especially for the second application. In
the Kansas plots, the investigators used different types of spray cards for the two sprays; however,
this was not the case in Oklahoma. The authors gave no other reasons that might explain the poor
results on some plots. Grand means for all treated plots were 33% and 21% of nominal for the fi rst
and second sprays, respectively. These values are in the same range as the values reported after
ground application in Kansas. Card deposits three metres into the fi eld edge had maximum (presum-
ably downwind) values ranging from less than 1 to 583% of fi eld deposits. Maximum edge deposits
exceeded average fi eld deposits in nine of 16 fi elds monitored.
Searchers found seven bird carcasses in the post-treatment periods in the carbofuran plots com-
pared with one bird in the pre-treatment 'sweeps' of the search areas. This represents an uncorrected