Place and Time - Sustainable Design: The Science of Sustainability and Green Engineering - page 182

and real-time advice. This is particularly important when confronted with a
complex or new problem. The mix of inputs from trusted mentors could make
for a solution very different from one where only handbooks are consulted.
For example, most professors are gratified when a former student or employee
contacts them about a specific problem or project. The mentor often has to
go back to his or her files or spend some time remembering similar situations,
but enjoys the challenge. This mentor-learner model also helps to ensure that
the knowledge and wisdom of this generation are passed on to the next (i.e.,
providing a way to preserve “corporate” memory in the ever-changing fields of
green design).
The sheer amount and complexity of data and information is enormous at
present and will continue to grow. In environmental engineering we have always
had to make decisions in the face of great amounts of uncertainty. Generally,
uncertainty comes from many sources. The data available to designers always
include some variability. The instruments used to gather the data will always
have internal variability (e.g., drift or effects from concentrations of chemicals
being tested). They will also have external variability, such as operator vari-
ability and temperature and pressure differences. Detection limits for chemi-
cals, for example, will vary from lab to lab and instrument to instrument. This
results from differences in standards, reagents, operators, instrument compo-
nents (e.g., wattage in lamps, types of mass spectrometry), and the standard
operating procedures at various labs. What we test is also highly variable. Air,
water, sediment, soil, and biota are dynamic systems. The water content in
each varies temporally. Sediment and soil organic contents vary slightly in the
near term (e.g., hours), but sometimes significantly over the long term (e.g.,
seasons, years).
The measurements that we take are often not quite as “direct” as we may like
to think. And even if data are straightforward to those of us who are technically
savvy, a lot of what scientists and engineers do does not always seem logical to a
broader audience. Thus, explaining the meaning of data can be very challenging.
That is due, in part, to the incompleteness of our understanding of the methods
used to gather data. Even well-established techniques such as chromatography
have built-in uncertainties. Since accuracy is how close we are to the “true
value” or reality, our instruments and other methods only provide data, not
information, and certainly not knowledge and wisdom. In chromatography, for
example, we are fairly certain that the peaks we are seeing represent the molecule
in question, but actually, depending on the detector, all we are seeing is the
number of carbon atoms (e.g., flame ionization detection) or the mass/charge
ratios of molecular fragments (e.g., mass spectrometry), not the molecule itself.
Add to this, instrument and operator uncertainties and one can see that even the
more accepted scientific approaches are biased and inaccurate, let alone approaches
such as mathematical modeling, where assumptions about initial and boundary