Environmental Engineering Reference
In-Depth Information
nomically valuable deciduous and coniferous trees. One solution to this dev-
astating environmental problem might be to aerially spray a pesticide (which
has been a practice in some parts of the region). Suppose a certain pesticide
has been chosen after clinical laboratory trials have demonstrated its effec-
tiveness against the caterpillar pest.
There are, however, two problems with applying the pesticide based on
knowledge acquired from the clinical studies. First, a clinical trial cannot
guarantee against confounding random environmental effects when the pes-
ticide is applied to a whole ecosystem. More importantly, many pesticides
are broad-spectrum agents, meaning that they kill more than the focal pest
species. Many pesticides have the potential to kill a whole host of butterflies
and moths that may offer important beneficial services in natural ecosystems
such as pollinating plants or being an important food base for many song-
bird species.Thus, a large-scale aerial application of the pesticide has the po-
tential benefit of arresting the pest species outbreak. But the potential cost
of the application is damage to other important components of the north-
ern forest ecosystem. Suppose that despite vocal public concerns, policy
makers felt that the risk was worth the market economic benefit of healthy
trees and thus called for the implementation of large-scale aerial spraying.
Suppose however, that management conducted a single, large-scale applica-
tion of the pesticide. In so doing, it missed the opportunity to treat this as
an experiment with a treatment area and control area and follow-up mon-
itoring of species and ecosystem responses to the pesticide.
Suppose that the application of the pesticide was confounded by a large-
scale, unavoidable random environmental effect after it was applied. For ex-
ample, an unusually cool autumn and a prolonged rainy and cold spring
would be sufficient to cause a natural downturn in the pest species popu-
lation via a large-scale reproductive failure. If such a random environmen-
tal effect occurred, it would swamp out the pesticide effect leading to a high
degree of uncertainty about the true effectiveness of the aerial spraying pro-
gram.The significance of this uncertainty is even more striking when we
consider it in light of the potential ancillary damage that the pesticide ap-
plication may have caused to other ecosystem components.The absence of
a control in this case meant that management missed a valuable opportu-
nity to collect important data on ecosystem-level responses to a large-scale
perturbation. Such information could have been used to devise more strate-
gic and targeted pesticide applications in the future. More importantly, if
management applied the same approach to any later outbreak and justified
