In the United States, millions of dollars are currently spent to monit
or water quality for a whole suite of organic compounds. However, resu
lts of several surveys conducted in the past decade indicate that only
a few pesticides occur in a small proportion of wells. Screening meth
ods based on historical evidence of contamination patterns and knowled
ge of the locales will have significant potential to reduce these cost
s and effectively identify contamination problems. In this paper, the
economics of utilizing two screening methods, sequential analysis and
sample compositing, in the design of monitoring strategies is captured
in the form of mathematical models and illustrated for a state-level
monitoring program. When the two methods are adopted, the total analyt
ical cost to conclusively identify contaminated wells in a network of
4,000 wells is shown to range from $12,500 to $1,575,000 depending on
the extent of contamination. In contrast, the total analytical cost of
a conventional program where all the wells in the network are sampled
and tested for a standard suite of pesticides at a cost of $250/sampl
e is one million dollars. Given such wide range in costs, it is pruden
t to incorporate the screening concepts presented in this paper in the
development of cost-effective monitoring programs.